T-cell factor 1 is a gatekeeper for T-cell specification in response to Notch signaling

Abstract

Although transcriptional programs associated with T-cell specification and commitment have been described, the functional hierarchy and the roles of key regulators in structuring/orchestrating these programs remain unclear. Activation of Notch signaling in uncommitted precursors by the thymic stroma initiates the T-cell differentiation program. One regulator first induced in these precursors is the DNA-binding protein T-cell factor 1 (Tcf-1), a T-cell-specific mediator of Wnt signaling. However, the specific contribution of Tcf-1 to early T-cell development and the signals inducing it in these cells remain unclear. Here we assign functional significance to Tcf-1 as a gatekeeper of T-cell fate and show that Tcf-1 is directly activated by Notch signals. Tcf-1 is required at the earliest phase of T-cell determination for progression beyond the early thymic progenitor stage. The global expression profile of Tcf-1-deficient progenitors indicates that basic processes of DNA metabolism are down-regulated in its absence, and the blocked T-cell progenitors become abortive and die by apoptosis. Our data thus add an important functional relationship to the roadmap of T-cell development.

Fig. 1.

Early thymocyte development in Tcf-1^−/− mice. Profiles of gated Tcf-1^−/− and control Lin⁻ DN thymocytes are shown. (A) CD44 versus CD25 profiles. The DN1–DN4 subsets are indicated. Numbers indicate the frequencies of each subset. (B Left) CD44 versus c-kit profiles show the electronic gating for pro-T cells. (Right) c-kit versus CD25 profiles of gated pro-T cells define the ETP, DN2a, and DN2b subsets as indicated. (C) Analysis of the DN1 subsets in Tcf-1^−/− and control mice. (Left) CD44 versus CD25 profiles show the gating of DN1 prothymocytes. (Right) CD24 versus c-kit profiles of gated DN1 cells define the DN1a (c-kit⁺CD24⁻), DN1b (c-kit⁺CD24⁺), DN1c (c-kit^loCD24⁺), DN1d (c-kit⁻CD24⁺), and DN1e (c-kit⁻CD24⁻) subsets. (D) Bar histograms of cell numbers in the indicated subsets of Tcf-1^−/− and Tcf-1^+/− mice (n = 4–6). (E) Histogram overlay of c-kit levels in gated DN1 thymocytes from the indicated mice. MFI, mean fluorescence intensity of c-kit.

Fig. 2.

BM and circulating blood progenitors in Tcf-1^−/− mice. (A) Analysis of gated Lin⁻ BM cells is shown. c-kit versus Sca-1 profiles show the gating for LSK progenitors. c-kit versus Flt3 profiles show the HSC, MPP, and LMPP subsets as indicated. Flt3 versus IL7Rα profiles show the gating for Lin⁻Flt3⁺IL7Rα⁺ cells, and c-kit versus Sca-1 plots depict the CLP. (B) Bar histograms show absolute numbers of HSCs, MPPs, LMPPs, and CLPs in the indicated mice (n = 3–6). Error bars are SD. Data are representative of three experiments. (C–E) Lymphocytes (0.6–1 × 10⁶) isolated from the blood of six mice for each group were stained for Lin as well as the indicated markers. (C) Gating strategy for blood LSK cells in the indicated mice. (D) Gating strategy for blood CLPs. (E) CTPs were gated as Lin⁻, HuCD25⁺. Histogram overlays show the surface profile for markers that define the CTP in the indicated mice. Unshaded histograms depict the CTPs, and shaded histograms depict an unstained negative control. Data are representative of two experiments.

Fig. 3.

T-cell potential of Tcf-1^−/− progenitors on OP9-DL1 cocultures. (A) Experimental scheme. Lin⁻ BM precursors or progenitor subsets, as defined in Fig. 1, were sorted from Tcf-1^−/− or WT mice. Equal numbers of Tcf-1^−/− or WT progenitors were mixed and cocultured on OP9-DL1 stroma for 10 d. (B) OP9-DL1 cultures of the indicated progenitors. CD45.1 versus forward-scatter (FSC) plots define the gating of Tcf-1^−/− (CD45.1⁻) versus WT (CD45.1⁺) cells. CD44 versus CD25 plots of the gated cells as indicated by arrows. Similar results were obtained in three independent experiments. (C) Apoptosis of Tcf-1^−/− and control progenitors in OP9-DL1 cultures was measured by annexin V staining. (Left) Histograms show annexin V staining of control (Upper) and Tcf-1^−/− (Lower) Lin⁻ BM progenitors cultured on OP9-DL1 cocultures for 10 d. (Right) Histogram bars show the average frequency of annexin V⁺ cells from the indicated mice in three cocultures. Error bars are SD.

Fig. 4.

Tcf-1^−/− progenitors are selectively defective in T-cell development. (A) Experimental scheme. Lin⁻ BM progenitors (5 × 10⁵) CD45.2⁺ sorted from Tcf-1^−/− were mixed with an equal number of WT CD45.1⁺ Lin⁻ BM progenitors and injected i.v. into lethally irradiated mice. After 10 wk, cells were harvested from BM, spleens, and thymi and analyzed. Progenitor subsets and mature lineages were gated as indicated to determine the fraction of CD45.2⁺ Tcf-1^−/− cells versus CD45.2⁻ WT cells in each population. (B) Histogram bars depict the relative contribution of Tcf-1^−/− versus WT progenitors in reconstituting the indicated progenitor subsets and mature lineages. Values represent the average of five independent mice. (C–E) Gating strategy for ETP (C), DN1–DN4 (D), and double-positive (DP) and single-positive subsets (E). Similar results were obtained with more than 10 reconstituted mice in three independent experiments.

Fig. 5.

Comparison of the c-kit+ DN1 thymocytes subset in Tcf-1^−/− versus WT mice. (A Left) CD44 versus CD25 plots depict the gating of DN1 cells in Tcf-1^+/− versus Tcf-1^−/− mice. CD44 versus c-kit plots are DN1 cells from Tcf-1^+/− and Tcf-1^−/− mice as indicated. The gate in these plots depicts Lin⁻CD44⁺c-kit⁺CD25⁻ (ETP) cells analyzed for apoptosis proliferation and global gene expression. (Right) Annexin⁺ and BrdU histograms depict the fraction of apoptotic and proliferating ETPs, respectively, in the indicated mice (Materials and Methods). (B) Gene ontology analysis of up- and down-regulated genes in the indicated processes in Tcf-1^−/− versus Tcf-1^+/− ETPs. Scale shows fold enrichment. (C) Heat map showing the mean expression of genes representative of the ETP/T-cell signature from the indicated mice. Means are the average of three biological replicates.

Fig. 6.

Tcf-1 is the target of Notch1. (A) Visualization of ChIP-Seq data in the Tcf-1 (Tcf7) locus. Notch-1 and CSL ChIP-Seq data are from the T6E cell line, and Tcf-1 ChIP-Seq data are from WT thymocytes. Mammalian conservation (conservation) is shown under the data tracks. Magnification of the indicated area strongly bound by Notch-1, CSL, and Tcf-1 is shown (∼31.5 kb upstream of the Tcf-1 transcription start site). This region contains conserved CSL and Tcf binding sites as indicated. (B) EMSA with nuclear extracts from the 1F9 T-cell acute lymphoblastic leukemia (T-ALL) cell line (lanes 2–4) or purified CSL (lanes 5–7). Probes are from the conserved CSL binding site in the putative Tcf-1 enhancer region, starting at position 52127891 on chromosome 11 (mm9). (C) Heat map depicting the expression of the indicated genes in LSK BM progenitors sorted before or after MxCre-mediated induction of a dormant oncogenic Notch1-IC (N1-IC⁺). Fold change of expression between N1-IC⁺ and WT LSK samples is shown.