Published online before print October 2, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Immunology Section, Lund University, and
Hematopoietic Stem Cell Laboratory, Lund Strategic Research Center for Stem Cell Biology and Cell Therapy, Lund University Hospital, Lund, Sweden;
Department of Clinical Sciences and Department of Pediatrics, Cellular Autoimmunity Unit, Malmö University Hospital, Malmö, Sweden; and
MRC Centre for Immune Regulation, Institute of Biomedical Research, University of Birmingham, Birmingham, United Kingdom
2 Correspondence: Immunology Section, Lund University, BMCI13, Lund, Sweden. E-mail: william.agace{at}med.lu.se
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-like 1 stromal cells. In contrast, early thymic progenitor and DN2 thymocyte numbers were reduced in the thymus of adult CCR9–/– mice. In fetal thymic organ cultures (FTOC), CCR9–/– DN1 cells were as efficient as WT DN1 cells in generating double-positive (DP) thymocytes; however, under competitive FTOC, CCR9–/– DP cell numbers were reduced significantly. Similarly, following intrathymic injection into sublethally irradiated recipients, CCR9–/– DN cells were out-competed by WT DN cells in generating DP thymocytes. Finally, in competitive reaggregation thymic organ cultures, CCR9–/– preselection DP thymocytes were disadvantaged significantly in their ability to generate CD4 single-positive (SP) thymocytes, a finding that correlated with a reduced ability to form TCR-MHC-dependent conjugates with thymic epithelial cells. Together, these results highlight a role for CCR9 at several stages of adult thymopoiesis: in hematopoietic progenitor seeding of the thymus, in the DN-DP thymocyte transition, and in the generation of CD4 SP thymocytes.
Key Words: chemokines • thymus seeding • T cell development
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
gene [2
]. Productive TCR rearrangement results in expression of TCRβ on the thymocyte surface, which is tested for its capacity to interact with the endogenous MHC. Positively selected DP thymocytes, which have received survival signals through the TCR, differentiate into CD4+ or CD8+ [single-positive (SP)] thymocytes as they reach the inner cortex, and these cells finally enter the medulla, where they undergo negative selection and terminal differentiation into mature SP cells. Such organized migration of developing thymocytes through distinct thymic niches promotes timely interactions with cortical and medullary thymic epithelial cells (cTEC and mTEC, respectively) and medullary dendritic cells (mDCs) and is believed to play a central role for normal thymocyte development (for reviews, see refs. [1
, 3
]). The chemokine CCL25 is expressed in the normal thymus by cTEC, mTEC, and mDCs [4 5 6 7 ]. Its receptor CCR9 is expressed on subsets of thymocytes and is tightly regulated during thymocyte development. CCR9 is expressed on a subset of DN thymocytes, and expression is up-regulated dramatically on DP cells following pre-TCR signaling [8 9 10 11 ]. CCR9 is subsequently down-regulated, as DP thymocytes differentiate into mature SP CD4+ cells, but is maintained in the mouse on SP CD8+cells [8 , 12 ]. Despite high levels of CCL25/CCR9 in the thymus, adult CCR9–/– mice display normal thymic cellularity and thymocyte composition [13 14 15 ]. Nevertheless, in competitive BM transfer experiments into RAG–/– [14 ] or irradiated wild-type (WT) mice [10 , 16 ], CCR9–/– cells were shown to have a severe disadvantage prior to or at the DN3 and early thymic progenitor (ETP; c-kit+CD25–CD44+CD4lo) stage of development, respectively. This reduced competitiveness of CCR9–/– BM cells in the generation of DN thymocytes has been suggested to reflect defects in thymocyte progenitor generation within the BM [10 ] and/or seeding of the adult thymus [14 , 16 ]. In addition, CCR9–/– DN2/3 thymocytes are localized throughout the cortex rather than in the SCR, in particular in young (12- to 14-day-old) mice [13 ], raising the possibility that CCR9 plays a role in intrathymic T cell development. Nevertheless, under noncompetitive conditions, this relocation of DN2/3 thymocytes appears not to affect subsequent thymocyte development [13 ].
To gain further insight into the role of CCR9 in adult thymopoiesis, we compared the competitiveness of CCR9–/– and WT cells in in vivo and in vitro transfer models of thymocyte development. Our results indicate that CCR9 is involved at multiple stages of thymocyte development, including the entry of T cell precursors to the adult thymus, the transition of DN thymocytes into DP cells, and in the generation of mature CD4 SP thymocytes.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antibodies
Anti-FcRII/III (2.4G2) hybridoma was obtained from American Type Culture Collection (Manassas, VA, USA); anti-CD3 (145-2C11), anti-CD4 (RM4-5), anti-CD4 (H129.19), anti-CD5 (53-7.3), anti-CD8 (53-6.7), anti-CD8β (53-5.8), anti-CD11b (M1/70), anti-CD11c (HL3), anti-CD25 (7D4), anti-CD25 (PC61), anti-CD34 (RAM34), anti-CD44 (IM7), anti-CD45.1 (A20), anti-CD45.2 (104), anti-TCRβ (H57-597), anti-IL-7R (A7R34), anti-TCR
(GL3), anti-B220 (RA3-6B2), anti-NK1.1 (PK136), anti-Gr-1 (RB6-8C5), anti-Ter119 (Ter-119), anti-Sca-1 (E13-161.7), anti-c-kit (2B8), rat IgG1 (R3-34), rat IgG2a (R35-95), rat IgG2b (A95-1), and fluorochrome-conjugated streptavidin were purchased from BD Biosciences (Stockholm, Sweden); anti-Sca-1 (D7) from eBiosciences (San Diego, CA, USA); monoclonal anti-CCR9 (7E7) kindly provided by Dr. Oliver Pabst (Hannover Medical School, Germany); tricolor (TC)-conjugated goat anti-rat IgG and PE-Cy5-conjugated sheep anti-rat from Caltag (Buckingham, UK); anti-rat IgG from Jackson ImmunoResearch Laboratories (West Grove, PA, USA); rat serum and 7-aminoactinomycin D from Sigma-Aldrich (Stockholm, Sweden); and TO-PRO-1 iodide from Molecular Probes (Eugene, OR, USA).
Cell isolation and flow cytometry analysis
BM, splenic, and mesenteric lymph node cells and thymocytes were isolated according to standard protocols.
BM cell staining
BM cells were incubated in a cocktail of predetermined optimal concentrations of Lin antibodies to B220, Gr-1, CD11b, CD8, CD5, CD4, and Ter119, followed by a TC-conjugated goat anti-rat antibody. After washing, cells were resuspended in Fc block followed by addition of antibody cocktails for lymphoid-primed multipotent progenitors (LMPP)/Lin– Sca-1+c-kit+ [LSK; Sca-1 FITC, fetal liver tyrosine kinase 3 (Flt3) PE, c-kit allophycocyanin (APC)] or common lymphoid progenitors (CLP; Sca-1 FITC, IL-7R PE, CD3 tandem conjugate, B220 tandem conjugate, c-kit APC).
Thymocyte staining
For ETP analysis, thymocytes were stained with a FITC-conjugated lineage cocktail (anti-CD8, CD8β, TCRβ, TCR, NK1.1, CD11b, Gr-1, and B220), anti-c-kit APC, and biotinylated anti-CD25, washed and incubated with Pacific blue-conjugated streptavidin. For some experiments, anti-CD44 Alexa-700 and anti-CD45 PE-Cy7 were added as additional markers. For DN subset analysis, thymocytes were incubated with FITC-conjugated lineage cocktail (anti-TCR, NK1.1, CD11b, Gr-1, and B220), anti-CD45 PE-Cy7, anti-CD4 APC, anti-CD8 PE, anti-CD44 Alexa-700, and biotinylated anti-CD25, washed and incubated with Pacific blue-conjugated streptavidin. For analysis of DN cells from hematopoietic stem cell (HSC) transfers, thymocytes were stained with a lineage cocktail (anti-CD3, -CD4, -CD5, -CD8, and -CD11b, B220, Ter119, and Gr-1), directly conjugated anti-CD44, anti-CD25, anti-CD45.1, and anti-CD45.2 antibody.
HSC adoptive transfers
LSK CD34– (long-term HSC) were isolated and purified (>98% pure) as described previously [17
18
19
20
]. C57BL/6.CD45.1+CD45.2+ mice were lethally irradiated (950 rad) and i.v.-injected with 250 µl PBS containing 200 CCR9–/– (CD45.2+), 200 WT C57BL/6.CD45.1 HSC, and 200,000 CD45.1+CD45.2+ unfractionated BM cells.
OP9--like 1 (OP9-DL1)/LMPP cultures
LSKFlt3hi (highest 25%) LMMP were purified (>97% pure) as described previously [21
, 22
]. The OP9-DL1 cell line (kindly provided by Dr. Anna Cumano, Institut Pasteur, Paris, France, and Dr. Juan-Carlos Zuniga-Pflucker, University of Toronto, Toronto, Canada) was maintained as described [23
]. Cells were trypsinized and prepared at a density of 2 x 104 cells/ml. Single or 10 LMPP were sorted directly onto 48-well plates containing OP9-DL1 cells [supplemented with 25 ng/ml Flt3 ligand (FL)] by a single-cell depositor unit on FACSDiVa (BD Biosciences). In some experiments, 10 LMPP from WT (CD45.1) mice and 10 CCR9–/– (CD45.2) LMPP were sorted directly into the same well. The cocultures were kept in complete OptiMEM medium containing L-glutamine (Invitrogen, Carlsbad, CA, USA), supplemented with FCS (Gibco, Paisley, Scotland) and 25 ng/ml FL for 28 days. Once a week, cytokines were replenished by adding 0.5 ml fresh medium. Clones were analyzed by flow cytometry at 28 days for T cell-committed progenitor analysis (CD45.1, CD45.2, CD4+CD8+), as described previously [21
].
Intrathymic injection
Sublethally irradiated C57BL6.CD45.1+CD45.2+ mice were injected intrathymically as described previously [24
]. Briefly, mice were anesthetized with isofluoran, and a small (1–2 cm) midline incision was made in the skin overlying the lower cervical and upper thoracic region. The upper third of the sternum was then bisected longitudinally to expose the thymus. FACS-sorted WT (Ly5.1+) and CCR9–/– DN (Ly5.2+) cells were mixed at a 1:1 ratio, and a total of 20,000 cells was injected in a total volume of 10 µl into each thymus lobe. Mice were killed 14 days later, and the CCR9–/–:WT donor ratio with the DP thymocyte population was determined by flow cytometry.
Fetal thymic organ cultures (FTOC)
Isolation of CD44+25– DN1 precursors
Thymocytes were incubated with FITC-conjugated antibodies to CD4 (clone GK1.5), CD8 (clone 53-6.7), and CD3 (clone 145.2C11), and positive cells were depleted using anti-FITC MicroBeads and MidiMACS depletion columns (Miltenyi Biotec, Auburn, CA, USA). From the remaining cells, CD44+25– DN1 cells were isolated by depletion using anti-CD25-coated Dynabeads (Dynal, Great Neck, NY, USA) and positive selection using anti-CD44-coated Dynabeads and Detachabead, according to the manufacturer’s instructions. Purities were typically 98% or greater. In some cases, WT and CCR9–/– DN1 cells were labeled with the fluorescent dyes PKH26 (Sigma-Aldrich) or CFSE (Molecular Probes), respectively [25
].
Hanging drop recolonization assay
Individual 1.35 mM 2-deoxyguanosine (dGuo)-treated FTOC [23
] were cultured in Terasaki plates with equal numbers of labeled WT and CCR9–/– DN1 cells (35,000 cells each). The thymus-colonizing ability of WT and CCR9–/– cells was analyzed after overnight culture by disaggregating individual lobes [25
], followed by flow cytometry to detect labeled cells. Long-term developmental potential of introduced cells was analyzed by transferring lobes to normal organ culture conditions, which were analyzed at the indicated time-points by flow cytometry.
Positive-selection reaggregation thymus organ cultures (RTOC)
Preselection CD4+8+69– [25
] thymocytes was isolated from adult WT (CD45.1+) and CCR9–/– (CD45.2+) thymuses by MoFlo cell sorting at a purity of >99%. Equal numbers of WT and CCR9–/– thymocytes were mixed together with dGuo-treated thymic stromal cell suspensions, obtained from disaggregated 2-deoxyguanosine-treated FTOC (
95% thymic epithelial cells; the remainder thymic fibroblasts), and used to form RTOC, as described [25
]. After 5 days, cultures were teased apart, and recovered thymocytes were analyzed for expression of CD4, CD8, and CD45.1.
Thymocyte-epithelial cell conjugate assays
Preselection CD4+8+69– thymocytes were prepared from adult WT and CCR9–/– adult thymus [23
], and WT and CCR9–/– cells were fluorescently labeled with DiD (Molecular Probes) and PKH26 (Sigma-Aldrich), according to the manufacturers’ instructions. Thymic epithelial cells, obtained by disaggregating dGuo-treated thymic organ cultures as described above, were labeled with CFSE (Molecular Probes). Equal numbers of freshly labeled thymic epithelial cells were mixed with a 1:1 mixture of WT/CCR9–/– thymocytes and centrifuged to form a cell pellet. Analysis of conjugate formation was performed at the indicated times using a BD-LSR (Becton Dickinson, San Jose, CA, USA) flow cytometer, as described [23
].
Statistics
Statistical analysis was performed using paired and unpaired Student’s t-test when appropriate.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
+, and TCR+ T cells derived from WT HSC was significantly greater than the number derived from CCR9–/– HSC in the spleen of recipient animals, and the number of B cells (B220) and DC (CD11c) derived from WT and CCR9–/– HSC was similar (data not shown). The percentage of DN, DP, and CD8/CD4 SP thymocytes derived from WT HSC was also significantly greater compared with CCR9–/– HSC (Fig. 1A
). Further examination of the DN thymocyte subset demonstrated that WT cells out-competed CCR9–/– cells at the DN1 stage of development (Fig. 1A)
. Together, these results confirm previous findings with whole BM-competitive transfers [10
, 14
, 16
] and suggest a role for CCR9 in the earliest stages of T cell development.
 View larger version (26K): [in a new window] |
Figure 1. CCR9–/– HSC are competitively disadvantaged in their ability to generate thymocytes, and adult CCR9–/– mice display reduced numbers of early DN thymocytes. (A) CCR9–/– (CD45.2+) and WT (CD45.1+) HSC were transferred together with support BM (CD45.1+CD45.2+) into lethally irradiated mice. The percentage of WT (solid bars) and CCR9–/– (open bars) cells within the indicated cell population in the thymus was determined 6 months after HSC transfer by flow cytometry. Results are from one representative experiment of three performed and are the means (±SEM) of seven to 11 mice per group; **, P < 0.01; ***, P < 0.001, compared with WT cells. (B) ETP are reduced in CCR9–/– mice. ETP were assessed in adult (8–10 weeks) CCR9–/– (open bar) and WT (filled bar) mice by flow cytometry. Representative stainings are shown in Supplemental Figure 1A. Mean (SD) of 11 mice/group combined from three separate experiments; ***, P < 0.001. (C) ETP express CCR9. WT and CCR9–/– thymic ETP Linlo (anti-CD8, CD8β, TCRβ, TCR, NK1.1, CD11b, Gr-1, B220, and CD25) c-kithi cells were stained with anti-CCR9 (solid line) or isotype control antibody (dashed line). Results are from one representative staining of four performed. (D) DN2 but not DN3 or DN4 thymocytes is reduced in adult (8–10 weeks) CCR9–/– mice. Representative stainings are shown in Supplementary Figure 1B. Mean (SEM) of 15 mice/group from four experiments; **, P < 0.01.
|
BM T cell progenitor development is normal in CCR9–/– mice
To assess whether the reduced number of ETP in CCR9–/– mice may be a result of reduced T cell progenitor generation in CCR9–/– BM, we examined HSC populations in the BM of CCR9–/– and WT mice (Fig. 2
and Supplementary Fig. 2). The total number of BM LSK and CLP (Lin– IL-7RhiSca-1loc-kitlo) was similar in WT and CCR9–/– mice (Fig. 2A)
. In addition, LSK cells that express high levels of Flt3, termed LMPP and displaying potent lymphoid progenitor cell potential [21
], were found in the same numbers in WT and CCR9–/– BM (Fig. 2A)
. As LMPP represent a likely source of T cell progenitors in the BM, the potential of CCR9–/– and WT LMPP to generate T cells was assessed in OP9-DL1 stromal cell cultures. LMPP (single or 10 cells) from CCR9–/– (Ly5.2) and WT (Ly5.1) mice were sorted onto confluent OP9-DL1 stromal cells, and their ability to generate CD4+CD8+ DP T cells was assessed 28 days later by flow cytometry. The cloning efficiency and the proportion of clones generating DP T cells in wells seeded with single WT and CCR9–/– LMPP were similar (Fig. 2B)
. In addition, seeding of 10 CCR9–/– or WT LMMP/well resulted in a similar percentage of wells containing DP T cells (Fig. 2B)
. Moreover, CCR9–/– LMPP appeared not to be disadvantaged compared with their WT counterparts in generating T cell clones in competitive OP9-DL1 cultures (Fig. 2C)
. Together, these results suggest that the reduction in ETP in adult CCR9–/– mice and disadvantage of CCR9–/– HSC in generating early DN thymocytes may be a result of reduced T cell progenitor seeding of the adult thymus rather than a reduced T cell progenitor potential of CCR9–/– BM.
 View larger version (40K): [in a new window] |
Figure 2. CCR9–/– BM contains normal numbers of LSK, CLP, and LMPP and has normal T cell progenitor potential. (A) BM LSK, CLP, and LMPP numbers in adult (8–10 weeks) CCR9–/– (open bars) and WT (filled bars) mice were assessed by flow cytometry. Representative stainings are shown in Supplementary Figure 2. Mean (SEM) of eight mice/group from two experiments. (B) Single or 10 LMPP from CCR9–/– and WT mice were sorted directly onto OP9-DL1 cells, or (C) 10 LMPP from WT (CD45.1+) mice and 10 CCR9–/– (CD45.2+) LMPP were sorted directly into the same well with OP9-DL1 cells, and their ability to generate CD4+CD8+ DP thymocytes was assessed 4 weeks later by flow cytometry.
|
 View larger version (37K): [in a new window] |
Figure 3. CCR9–/– and WT DN1 thymocytes are equally efficient at generating DP thymocytes in FTOC. (A) WT and CCR9–/– DN1 thymocyte development in FTOC cultures under noncompetitive conditions. DN1 thymocytes from 3- to 5-week-old CCR9–/– (open bars) and WT (filled bars) mice were added separately to dGuo-treated E15 thymic lobes. Thymocytes were recovered from FTOC cultures after 12 days of culture, and the number of DP (CD4+CD8+), CD8 SP, and CD4 SP cells was determined by flow cytometry analysis. Mean (SD) of four mice/group. **, P < 0.01. (B) Representative flow cytometry staining from CCR9–/– and WT FTOC cultures after 12 days of culture. Note the reduction in CD4 SP cells in FTOC receiving CCR9–/– DN1 cells.
|
 View larger version (42K): [in a new window] |
Figure 4. CCR9–/– thymocytes are disadvantaged in their ability to generate DP thymocytes under competitive conditions. (A and B) CCR9–/– DN1 cells enter E15 thymic lobes as efficiently as their WT counterparts. DN1 thymocytes from 3- to 5-week-old CCR9–/– and WT mice were labeled with PKH26 and CFSE, respectively, mixed at a 1:1 ratio, and added to dGuo-treated E15 thymic lobes, which were disaggregated after overnight culture, and the number of PKH26+ and CFSE+ cells determined by flow cytometry. Representative flow cytometry plots of cells found outside and inside the lobe. (B) Results are the ratio of WT:CCR9–/– cells from inside the lobe and are normalized to the WT:CCR9–/– ratio outside the lobe. Mean (SD) of six thymic lobes from two experiments. (C and D) Competitive WT and CCR9–/– DN1 thymocyte development in FTOC cultures. DN1 thymocytes from 3- to 5-week-old CCR9–/– (Ly5.2) and WT (Ly5.1) mice were mixed together at a 1:1 ratio and added together to dGuo-treated E15 thymic lobes. The number of CCR9–/– and WT DP and SP thymocytes was assessed 12 days later by flow cytometry. (C) Combined mean (SD) from seven thymic lobes from two experiments. **, P < 0.01; ***, P < 0.001. (D) Representative flow cytometry staining demonstrating the ratio of CCR9–/– and WT cells in the DP, CD4 SP, and CD8 SP thymocyte compartment. (E) Competitive DN thymocyte development after intrathymic injection. DN thymocytes from adult (10–12 weeks) CCR9–/– (CD45.2+) and WT (CD45.1+) mice were injected at a 1:1 ratio into the thymus of sublethally irradiated recipient (CD45.2+ CD45.1+) mice. The WT:CCR9–/– ratio in the DP thymocyte compartment of individual mice ( ) was assessed 14 days later by flow cytometry. Results are from 11 mice from two separate experiments. (F) Flow cytometry staining for CCR9–/– (CD45.2+ CD45.1–) and WT (CD45.1+ CD45.2–) cells in the DP thymocyte compartment.
|
cells, an alternative explanation for this finding is that CCR9 is involved in DP thymocyte transition to mature SP cells. To assess this possibility more definitively, preselection CD4+8+69– thymocytes were isolated from CCR9–/– and WT mice, mixed at a 1:1 ratio, and placed in RTOC. In these experiments, the total number of CCR9–/– DP thymocytes was slightly but significantly higher than that of WT DP cells (Fig. 5A
and 5B
), indicating that CCR9 does not play an important role in DP thymocyte survival in these cultures. Nevertheless, although CCR9–/– CD4+8+69– thymocytes were slightly more efficient at generating CD8 SP thymocytes compared with WT cells, they were significantly less efficient at generating CD4 SP thymocytes (Fig. 5A
and 5B)
. CCR9–/– CD4 SP and CD8 SP thymocytes expressed high levels of CD3 similar to WT cells, confirming that CCR9–/– DP thymocytes were disadvantaged in generating mature CD4 SP thymocytes (Fig. 5C)
. The duration of antigen receptor signaling plays a central role in the development of CD4 and CD8 SP thymocytes from DP precursors [26
, 27
], and prolonged signal duration is required for the development of CD4 SP cells [27
]. We therefore hypothesized that the selective deficiency in the ability of CCR9–/– DP thymocytes to generate CD4 SP cells may be a result of the reduced ability of these cells to form stable interactions with thymic epithelial cells. To address this possibility, we analyzed initial stages of positive selection using thymocyte-epithelial cell conjugate assays, in which cell-cell contact is dependent on TCR-MHC interactions. CCR9–/– and WT preselection DP thymocytes were labeled with the fluorescent dyes DiD and PKH26, respectively, and mixed at a 1:1 ratio, and their ability to interact with CFSE-labeled thymic epithelial cells was assessed in conjugate assays [25
]. Results from this analysis demonstrate that CCR9–/– CD4+8+ thymocytes are disadvantaged significantly compared with their WT counterparts in generating stable conjugates with thymic epithelial cells (Fig. 5D
and 5E)
. These results suggest that CCR9-mediated interactions with thymic epithelial cells are involved during initial stages of thymocyte positive selection.
 View larger version (50K): [in a new window] |
Figure 5. CCR9–/– DP thymocytes are disadvantaged in their ability to generate CD4 SP cells in competitive RTOC. (A) Preselection CD4+8+69– DP thymocytes were isolated from 4- to 6-week-old CCR9–/– (CD45.2+) and WT (CD45.1+) mice, mixed at a 1:1 ratio, and placed with dGuo-treated thymic stromal cells to form RTOC cultures. The total number of DP, CD4, and CD8 SP CCR9–/– (open bars) and WT (filled bars) thymocytes in RTOC cultures was assessed 5 days later by flow cytometry. Results are the mean (SD) of five RTOC from two separate experiments. n.s, Not significant; **, P < 0.01. (B) Representative flow cytometry analysis of CCR9–/– and WT thymocytes isolated from RTOC cultures. (C) CD3 expression on WT (shaded) and CCR9–/– (blank) DP, CD4 SP, and CD8 SP thymocytes from competitive RTOC cultures. (D and E) Kinetics of CCR9–/– and WT conjugate formation with thymic epithelial cells. (D) Flow cytometry results are from one representative conjugate assay. KO, Knockout. (E) Mean (SEM) conjugates formed between WT () or CCR9–/– ( ) CD69– preselection thymocytes and thymic epithelial cells. Data combined from five experiments. *, P < 0.05; **, P < 0.01.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CCR9–/– BM cells are disadvantaged in their ability to generate DN3 and ETP cells after competitive whole BM transfer into RAG–/– and WT mice, respectively [10 , 14 , 16 ], and these results were confirmed in the present study using competitive, purified HSC transfers. Furthermore, consistent with a recent publication [16 ], thymic ETP and DN2 cells but not DN3/4 populations were reduced significantly in adult CCR9–/– mice, indicating an important role for CCR9, even under noncompetitive conditions in the development of T cell progenitors in the BM and/or in hematopoietic cell seeding of the adult thymus. The observation that subsets of BM, MPP, and CLP express CCR9 [16 ] and that enhanced green fluorescent protein (EGFP)+ cells are found among BM, LSK, and CLP in CCR9-EGFP reporter mice [28 ] indicates that CCR9 may function in progenitor development within the BM. Indeed, the absence of CCR9 on ETP has been used to argue for a role for CCR9 in the generation of T cell precursors in the BM and not in thymic seeding [10 ]. Nevertheless, in contrast to the findings of Wurbel et al. [10 ], but consistent with two recent studies [16 , 29 ], we found that thymic ETP express CCR9. We demonstrate further that CCR9–/– BM contains normal numbers of HSC, CLP, LSK, and LMPP, and importantly that single CCR9–/– LMPP are equally capable of generating DP T cells on OP9-DL1 stromal cells as WT LMPP. Collectively, these results argue against a role for CCR9 in T cell progenitor development in the BM.
A role for CCR9 in T cell progenitor seeding of the adult thymus is at odds with the observation that adoptively transferred Lin– BM cells from CCR7/CCR9 double-KO mice are equally efficient at entering the adult thymus as their WT counterparts [30 ]. However, the use of total Lin– cells may have masked a role for CCR9 in thymic seeding of a minor but relevant population of thymocyte progenitors in this study. Indeed, i.v.-injected CCR9–/– BM cells were shown recently to populate the BM LSK population of mice with a similar efficiency as WT BM cells but were less efficient at generating thymic ETP. In contrast, intrathymically injected WT and CCR9–/– BM cells were equally efficient at generating ETP [16 ]. Although the identity of the T cell progenitors that seed the adult thymus under physiological conditions, is currently unclear [31 ], i.v.-injected CLP-2 cells, which have T cell precursor potential [32 ], require CCR9 for efficient entry into the thymus [33 ]. CLP-2 are unlikely to represent a thymic seeding population as they are not found in blood, however, Krueger and von Boehmer [29 ] recently detected a T lineage-committed progenitor in blood, which they termed circulating T cell progenitors, and these cells expressed high levels of CCR9. Together with our current findings that CCR9–/– BM has normal T lymphocyte potential, these results collectively suggest a role of CCR9 in hematopoietic progenitor seeding of the adult thymus.
CCR9 is expressed at low-to-intermediate levels on DN2 and DN3 thymocytes, respectively [9 , 10 ], and at high levels on DN4/DP thymocytes as a result of pre-TCR signal-induced CCR9 transcription at the DN3 stage [8 9 10 11 , 34 ], indicating a potential role for CCR9 in the development of DP thymocytes from their DN precursors. Nevertheless, adult CCR9–/– mice have normal numbers of DP thymocytes [13 14 15 ]. Consistent with these findings, we observed that CCR9–/– DN1 cells were as efficient as their WT counterparts at generating DP thymocytes in FTOC cultures under noncompetitive conditions. However, under competitive conditions, CCR9–/– DN1 thymocytes displayed a marked reduction in their ability to generate DP thymocytes. Furthermore, DN thymocytes from CCR9–/– mice were competitively disadvantaged compared with WT cells in their ability to generate DP thymocytes following intrathymic injection. In competitive OP9-DL 1 cultures, CCR9–/– and WT LMPP were equally efficient in generating DP thymocytes; however, OP9-DL1 cells do not appear to express CCL25 (unpublished observation). Collectively, these findings suggest that the advantage of WT over CCR9–/– cells is a result of microenvironmental factors, presumably as a result of competition for epithelial-derived CCL25. CCR9–/– DN2/3 thymocytes display aberrant localization throughout the thymic cortex, in particular, in younger (12- to 14-day-old) mice [13 ]. Thus, aberrant localization of CCR9–/– DN2/3 cells may result in developmental disadvantage compared with WT cells under competitive conditions. Of note, Uehara et al. [34 ] recently generated two transgenic mouse lines expressing CCR9 under control of the human CD2 promoter, and one of these lines showed a partial developmental block at the DN3 stage of development, as a result, at least in part, of a proliferative defect during the DN3-4 transition. Thus, the regulated expression of CCR9 also appears to be required for efficient development of DN cells into DP thymocytes. Although the compensatory mechanisms, which allow normal DP thymocyte development in CCR9–/– mice or in noncompetitive FTOC, remain to be determined, two potential candidate molecules are the chemokine receptors CCR7 (expressed on DN1-2 and a subpopulation of DN4 thymocytes) and CXCR4 (expressed on virtually all DN2/3/4/DP thymocytes) [9 , 35 ]. CCR7 and CXCR4 have demonstrated roles in adult intrathymic DN thymocyte development [9 , 35 ], and further studies using combinatorial KO should help determine any overlapping function for these receptors in the development of DP thymocytes.
CCR9 is expressed at high levels on DP thymocytes, on immature (CD69+CD62L–) but not mature (CD69–CD62L+) SP CD4 thymocytes, and on mature and immature SP CD8 thymocytes [8 , 9 , 36 ]. CCR9–/– mice have normal numbers of DP and SP thymocytes [14 , 15 ], indicating that CCR9 is not involved in the DP-to-SP transition or that there are compensatory mechanisms in place to ensure normal SP development. Here, we demonstrate that CCR9–/– DN1 cells generate similar numbers of DP and CD8 SP thymocytes as WT DN1 cells in noncompetitive FTOC. Furthermore, in competitive RTOC, CCR9–/– DP and SP CD8 cells were increased compared with WT, although the differences in CD8 SP numbers did not reach significance. Thus, CCR9 does not appear to play a role in the maintenance of DP thymocytes or in the development of CD8 SP cells in these cultures. In contrast, in noncompetitive FTOC cultures, we found that CCR9–/– DN1 cells showed a slight but significant disadvantage in generating SP CD4 cells. More strikingly, in competitive RTOC CCR9–/– preselection DP thymocytes were selectively disadvantaged in their ability to generate CD4 SP cells. This finding suggested to us that under competitive conditions, CCR9–/– DP thymocytes in RTOC received efficient signals for CD8 SP but not CD4 SP thymocyte development. Yasutomo et al. [27 ], using a modified, two-stage reaggregate culture system, demonstrated recently that the duration of antigen receptor signaling determines CD4 versus CD8 T cell lineage fate. Thus, short TCR signaling in DP cells results in the preferential generation of CD8 SP thymocytes, and extended signaling results in the preferential generation of CD4 SP thymocytes. In competitive aggregation studies, we found that CCR9–/– preselection thymocytes are compromised in their ability to form conjugates with thymic epithelial cells. Together, these results suggest an important role for CCR9 in regulating DP thymocyte interactions with thymic epithelial cells and provide a likely molecular explanation for the selective reduction of CCR9–/– CD4 SP cells in competitive RTOC. We hypothesize that compensatory adhesive interactions between DP thymocytes and thymic epithelia, potentially mediated by other chemokines, allow CD4 SP development to proceed normally in adult CCR9–/– mice.
In conclusion, using competitive transfer models, the current study provides new insights into the role of CCR9 at multiple stages of adult thymopoiesis. The future challenge will be to unravel the complex compensatory signaling networks and overlapping functional events that allow thymocyte development to proceed relatively unperturbed in animals deficient in this receptor.
Received June 21, 2007; revised September 5, 2007; accepted September 5, 2007.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
TCR+ thymocytes preferentially respond to CCL25 J. Immunol. 168,134-142(+) gut intraepithelial lymphocytes Blood 98,2626-2632-like-1 in vitro Immunity 17,749-756[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
