CCR7 and CCR9 together recruit hematopoietic progenitors to the adult thymus

Abstract

T lymphopoiesis requires settling of the thymus by bone marrow-derived precursors throughout adult life. Progenitor entry into the thymus is selective, but the molecular basis of this selectivity is incompletely understood. The chemokine receptor CCR9 has been demonstrated to be important in this process. However, progenitors lacking CCR9 can still enter the thymus, suggesting a role for additional molecules. Here we report that the chemokine receptor CCR7 is also required for efficient thymic settling. CCR7 is selectively expressed on bone marrow progenitors previously shown to have the capacity to settle the thymus, and CCR7(-/-) progenitors are defective in settling the thymus. We further demonstrate that CCR7 sustains thymic settling in the absence of CCR9. Mice deficient for both CCR7 and CCR9 have severe reductions in the number of early thymic progenitors, and in competitive assays CCR7(-/-)CCR9(-/-) double knockout progenitors are almost completely restricted from thymic settling. However, these mice possess near-normal thymic cellularity. Compensatory expansion of intrathymic populations can account for at least a part of this recovery. Together our results illustrate the critical role of chemokine receptor signaling in thymic settling and help to clarify the cellular identity of the physiologic thymic settling progenitors.

Figure 1

Functional CCR7 is selectively expressed by bone marrow progenitors. (A) The indicated populations were sorted from WT mice and cDNA was prepared. Real-time PCR was then used to quantify the relative levels of CCR7 transcripts. Shown is the mean ± SEM for 3 independent experiments. (B) Bone marrow (BM) from wild-type (WT) and CCR7^−/− mice were analyzed by flow cytometry for CCR7 expression on the indicated populations. Numbers represent the percentage of WT cells in the indicated gate. (C) For chemotaxis assays, WT BM cells were added to transwells with the indicated chemokines in the bottom and/or top wells. After a 2-hour incubation, the migrated cells were collected and quantified by flow cytometry. Shown is the mean ± SEM for data pooled from 3 experiments. *P < .05 for the indicated condition compared with the no-chemokine control condition. (D) BM from WT mice was analyzed for the coexpression of CCR7 and CCR9. Shown are cells previously gated as LMPPs (left panel) and CLPs (right panel). The gates showing CCR7⁻CCR9⁺, CCR7⁺CCR9⁺, and CCR7⁺CCR9⁻ subpopulations were placed according to CCR7^−/− and CCR9^−/− controls (data not shown). (E) WT and CCR9^−/− BM were analyzed by flow cytometry for CCR7 and CCR9 coexpression. The gray histogram in the third column represents the CCR7^−/− control.

Figure 2

CCR7 mediates thymic settling by T-lineage progenitors. (A) Thymi from WT and CCR7^−/− mice were analyzed by flow cytometry to calculate absolute numbers of ETPs. Absolute numbers were obtained by multiplying the frequency of ETPs among live singlets by total thymic cellularity. Shown is the mean ± SEM for 21 WT (8 males, 13 females) and 18 CCR7^−/− (7 males, 11 females) mice all between the ages of 4 and 9 weeks. These data were pooled from 4 separate experiments, but for each experiment the mice were age and sex matched. *P < .05 compared with the WT control. (B) Mixed BM chimeras were generated using CD45.2 WT BM and CD45.1 WT BM mixtures as controls (top panel), CD45.2 CCR7^−/− BM and CD45.1 WT BM mixtures (middle panel), or CD45.2 CCR9^−/− BM and CD45.1 WT BM mixtures (bottom panel). Chimeras were analyzed by flow cytometry after 10 weeks using antibodies to CD45.1 and CD45.2 to determine donor chimerism. Shown is the mean CD45.2 donor chimerism ± SEM for each indicated population. *P < .05 and **P < .01 for the CD45.2 donor chimerism of the indicated population compared with HSC CD45.2 donor chimerism. (C) For adoptive transfer experiments, BM from either WT or CCR7^−/− mice (both CD45.2) was administered intravenously (left 2 panels) or intrathymically (right panel) to unirradiated WT recipient mice (CD45.1). After 2 weeks (intrathymic transfer) or 3 weeks (intravenous transfer), donor-derived cells were quantified by flow cytometry. Results show the mean ± SEM for 5 recipients in each group. *P < .05 compared with the WT control.

Figure 3

CCR7/CCR9 DKO mice have a severe ETP defect. (A) BM from WT and CCR7/CCR9 DKO mice were analyzed by flow cytometry to determine frequencies of the indicated progenitors. Shown is the mean ± SEM for 9 WT and 10 CCR7/CCR9 DKO mice. (B) Thymic cellularities from WT and CCR7/CCR9 DKO mice were calculated. Shown is the mean ± SEM. *P < .05 compared with the WT control. (C) WT and CCR7/CCR9 DKO thymi were analyzed by flow cytometry for thymic progenitors. Representative FACS profiles are shown. (D) Absolute numbers of thymic progenitor populations were calculated. Shown is the mean ± SEM. *P < .05 and **P < .01 compared with the corresponding WT controls.

Figure 4

CCR7/CCR9 DKO progenitors have an absolute defect in thymic settling in the competitive scenario. (A) Mixed BM chimeras were generated using CD45.2 WT BM and CD45.1 WT BM mixtures (top row) or CD45.2 CCR7/CCR9 DKO BM and CD45.1 WT BM mixtures (bottom row). Chimeric mice were analyzed by flow cytometry 10 weeks after BM transfer for donor chimerism of the indicated populations using antibodies to CD45.1 and CD45.2. Representative FACS plots of BM LSKs (left column), thymic ETPs (middle column), and thymic DPs (right column) are shown. (B) Shown is the mean CD45.2 donor chimerism ± SEM for each indicated population from control WT chimeras (top panel) and CCR7/CCR9 DKO chimeras (bottom panel) described in panel A. **P < .01 for the CD45.2 donor chimerism of the indicated population compared with HSC CD45.2 donor chimerism. (C) For adoptive transfer experiments, BM from either WT or CCR7/CCR9 DKO (both CD45.2) was administered intravenously (left 2 panels) or intrathymically (right 2 panels) into unirradiated WT recipient mice (CD45.1). After 2 weeks (intrathymic transfer) or 3 weeks (intravenous transfer), donor-derived thymocytes were quantified by flow cytometry. Results show the mean ± SEM for 5 recipients in each group. **P < .01 compared with the WT control.

Figure 5

Compensatory proliferation accounts for near-normal CCR7/CCR9 DKO thymic cellularity. (A) Sorted ETPs from WT or CCR7/CCR9 DKO mice were cultured on OP9-DL1 stromal layers. WT cells were cultured in triplicate wells, whereas CCR7/CCR9 DKO cells were cultured in a single well. After 2 weeks, cultured cells were analyzed by flow cytometry for expression of CD25 and Thy1. Representative FACS plots of cells previously gated to be CD45⁺ are shown. (B) Graphs show the relative numbers of cells obtained from cultures described in panel A. Shown is the mean ± SEM for WT and CCR7/CCR9 DKO cultures. (C) Lin⁻Flt3^hi cells (3 × 10³) sorted from WT BM (CD45.1) were intrathymically injected into WT or CCR7/CCR9 DKO recipients (both CD45.2). After 15 days, recipient thymi were analyzed for donor-derived DN3 cells (left column) or DP cells (right column). Shown are representative donor/host FACS plots. (D) Graphs show the total numbers of donor DN3 cells (left panel) and donor DP cells (right panel) found in recipient thymi for the experiment described in panel C. Shown is the mean ± SEM for 5 WT recipients and 3 CCR7/CCR9 DKO recipients. **P < .01 compared with WT controls.