Polycomb group (PcG) proteins play a role in the maintenance of cellular identity throughout many rounds of cell division through the regulation of gene expression. In this report we demonstrate that the loss of the PcG gene mel-18 impairs the expansion of the most immature T progenitor cells at a stage before the rearrangement of the TCR β-chain gene in vivo and in vitro. This impairment of these T progenitors appears to be associated with increased susceptibility to cell death. We also show that the expression of Hes-1, one of the target genes of the Notch signaling pathway, is drastically down-regulated in early T progenitors isolated from mel-18−/− mice. In addition, mel-18−/− T precursors could not maintain the Hes-1 expression induced by Delta-like-1 in monolayer culture. Collectively, these data indicate that mel-18 contributes to the maintenance of the active state of the Hes-1 gene as a cellular memory system, thereby supporting the expansion of early T progenitors.

The Polycomb group (PcG)3 genes were originally identified in Drosophila as a class of regulators responsible for maintaining homeotic gene expression by contributing to the cellular memory of somite identity throughout cell division. PcG genes are conserved from Drosophila to mammals, and their protein products have been reported to localize to the nucleus as multimeric protein complexes. These proteins epigenetically maintain the repressed state of target genes through the modification of chromatin structure. In Drosophila, at least two types of PcG complexes, each with different properties, can be distinguished: Polycomb repressive complex 1 (PRC1) and ESC-E(Z) (1, 2). The mel-18 gene is a mammalian homologue of the Drosophila posterior sex combs (Psc) gene, and its product is a member of a PcG protein complex that also contains M33, BMI-1, RAE-28, RING1A, and RING1B (3, 4). This mammalian complex is similar to PRC1 in Drosophila, which is able to competitively inhibit the chromatin remodeling complex, SWI/SNF, and interacts with sequence-specific, DNA-binding factors (Pipsquesk, Zeste, and GAGA) and histone deacetylase (5, 6, 7). This PRC1 is shown to collaborate with the ESC-E(Z) complex in regulation of gene expression through epigenetic modification of chromatin structure (8, 9). Although several studies have recently described the silencing mechanisms of PcG complexes, to our knowledge, it has rarely been reported that the PcG gene is practically required for the maintenance of gene expression in the mammalian cell differentiation system.

It is well known that PcG genes play a significant role in the regulation of lymphocyte differentiation (10, 11). Mice deficient in the individual components of the PRC1-like complex, mel-18, bmi-1, rae-28, and m33, display SCID (12, 13, 14, 15). Loss of function of bmi-1 causes a severe block of B cell development (12), and rae-28 deficiency reduces the generation of pre-B and immature B cells from fetal liver (FL) hemopoietic progenitors (15). In T cell development, bmi-1 mutant mice exhibited impaired thymocyte development at an immature stage (16). In mel-18 mutant mice, B cell maturation is arrested between the pro- and pre-B cell stages, and severe thymic atrophy is also observed (14). In mature resting B cells, mel-18 negatively regulates B cell receptor-induced proliferation through the down-regulation of the c-Myc/cdc25 cascade (17). Th2 cell differentiation is also impaired in mel-18 mutant mice, and mel-18 is involved in the induction of GATA-3 under Th2-skewed conditions (18). In contrast to the extensive analysis of the roles of mel-18 in both immature and mature B cell development as well as in mature T cell function, much remains to be investigated with regard to a potential role for mel-18 in thymocyte development. Therefore, we performed a comprehensive analysis of mel-18 function to clarify its role and the underlying mechanisms in the regulation of thymocyte development.

The thymus is the major site of T cell differentiation and maturation. Thymocytes can be divided into four main populations: CD4CD8 double negative (DN), CD4+CD8+ double positive (DP), and CD4+ or CD8+ single positive (CD4SP or CD8SP) cells. DN cells can be further divided into four subpopulations (DN1-DN4) according to their CD44 and CD25 expression patterns. CD44+CD25 (DN1) thymocytes represent the earliest T progenitors in the thymus. DN1 cells differentiate through a CD44+CD25+ (DN2) stage into CD44CD25+ (DN3) cells, in which the TCRβ rearrangement takes place. Those with a successful TCR rearrangement at the β locus receive a pre-TCR signal and differentiate through the CD44CD25 (DN4) stage into DP cells (19). During intrathymic T cell development, thymocytes are required to expand massively to create a highly diversified TCR repertoire. It has been proposed that two independent proliferative phases exist during T cell development. One is the period of expansion that occurs before the TCRβ gene rearrangement (pre-β proliferation), which occurs during the DN1 and DN2 stages (20). The second phase of expansion occurs after TCRβ gene rearrangement (post-β proliferation), which is stimulated by the pre-TCR signal (21, 22).

Several signaling pathways have been shown to be involved in the regulation of proliferation and differentiation of thymocytes. Notch signaling has recently drawn wide attention for its role in T cell development (23). Notch signaling is an evolutionarily conserved pathway that controls multiple cell fate decisions in various tissues throughout ontogeny. Notch-RBP-J signaling is essential in the cell fate decisions between T and B cell lineages (24, 25, 26), αβ T cell and γδ T cell lineages (27, 28, 29), and Th1 and Th2 cell differentiation (30). Furthermore, several studies have indicated that Notch-1 signaling regulates the proliferation of early thymocytes in addition to the cell lineage commitment. A stromal cell line forced to express DL1 can support T cell development to at least the DP stage as well as the expansion of T progenitors in monolayer culture (31). The Hes-1 gene, Notch-1 signaling target gene, was shown to be essential for the expansion of early T progenitor cells, but not for T/B cell lineage commitment. The expression level of Hes-1 is involved in supporting the cellularity of developing thymocytes (32, 33).

In this study we found that loss of the PcG gene mel-18 impairs the expansion of early T progenitor cells at a stage before rearrangement of the TCR β-chain gene in vivo and in vitro. We also observed that the Hes-1 gene was down-regulated in early T progenitors in mel-18−/− mice. Based on our analysis, mel-18 appears to be indispensable for the maintenance of the active state of the Hes-1 gene in proliferating T progenitors. We propose that mel-18 contributes to the expansion of early T progenitors through the maintenance of cellular memory of these proliferating cells.

The mel-18 deficient (mel-18−/−) mice were described previously (14), and animals used in this study were backcrossed to C57BL/6 >10 times. C57BL/6J (B6, Ly5.2) mice were obtained from CREA Japan, and B6 SJL Ptprca Pep3bBoyJ (B6 Ptprc, Ly5.1) mice were obtained from The Jackson Laboratory. All mice were kept in accordance with the laboratory animal science guidelines of Hiroshima University. For timed pregnancies, the day of vaginal plug was counted as day 0.5.

Single-cell suspensions from bone marrow, thymus, FL, and fetal blood (FB) were stained with mAbs and second reagents. The following mAbs were purchased from BD Pharmingen: anti-CD4-FITC, -PE, -allophycocyanin, and -biotin (RM4-5); anti-CD8α-PE, -allophycocyanin, and -biotin (53-6.7); anti-CD3ε-FITC, -allophycocyanin, and -biotin (145-2C11); anti-CD44-PE (IM7); anti-CD25-FITC and -allophycocyanin (7D4 and PC61); anti-c-Kit-allophycocyanin (2B8); anti-Sca-1-FITC (D7); anti-CD45R/B220-FITC and -biotin (RA3-6B2); anti-CD19-FITC and -biotin (1D3); anti-Mac-1-FITC and -biotin (M1/70); anti-Gr-1-FITC and -biotin (RB6-8C5); Ter119-biotin; anti-NK1.1-FITC (PK136); anti-CD45.2-FITC (104); and anti-CD45.1-PE (A20). Anti-CD127-PE and -biotin (A7R34) and Ter119-FITC were purchased from eBiosciences. Biotinylated Abs were revealed with streptavidin-CyChrome (BD Pharmingen). To analyze DN CD3 thymocytes, lineage marker-negative (Lin)ScaI+c-Kit+ cell (KSL), and common lymphoid progenitors in bone marrow and p-T cell in FL, all cells expressing lineage markers (CD4, CD8, CD3, B220, CD19, Mac-1, Gr-1, Ter119, and NK1.1) were gated out of the analyses. FACS analysis was performed on a FACSCalibur flow cytometer (BD Biosciences), and data were analyzed using CellQuest software. For cell sorting, all cells were stained with biotinylated lineage markers, bound to streptavidin-magnetic beads and depleted of lineage-positive cells using MACS separation column (Miltenyi Biotec). The lineage-negative cells were then stained with subsequent Abs as described above. Cells were subsequently sorted using a FACSVantage SE (BD Biosciences). Dead cells were removed from analysis and sorting by staining with propidium iodide (PI; Sigma-Aldrich). Reanalysis of the sorted cells indicated the purity to be >94% for each cell population. For experiments examining mRNA expression in sorted cells, 3 × 104 cells from each fraction were subjected to real-time PCR. Cells (1 × 103) from 14.5E DN1 and DN2 fractions were subjected to semiquantitative PCRs.

All liquid cultures were performed in RPMI 1640 (Invitrogen Life Technologies) plus 10% FCS for FTOC, 50 μM 2-ME (Nacalai Tesque), 1 mM sodium pyruvate (Invitrogen Life Technologies), 1× nonessential amino acid solution (Invitrogen Life Technologies), and antibiotics. For analysis of T cell progenitors, FTOC under high oxygen submersion (HOS) conditions was performed as previously described (34). In brief, sorted cells from mel-18−/− and mel-18+/− Ly5.2 14.5 days post coitum (dpc), or Ly5.1/5.2 14.5 dpc were plated at one or 25 cells/well onto 96-well, V-bottom culture plates (Nalge Nunc International) containing deoxyguanosine-treated Ly5.1/5.2(F1) 15.5 dpc, or mel-18−/− and mel-18+/− 15.5 dpc FT lobes and cultured in medium. Stem cell factor (SCF; 10 ng/ml; R&D Systems) and 50 U/ml IL-7 (Genzyme) were added to the medium. Cultures were scored on day 10 or 14, and cells were analyzed by flow cytometry.

Semiquantitative RT-PCR was performed on cDNAs obtained from 1000 sorted cells of DN1 (CD44+CD25c-Kit+) and DN2 (CD44+CD25+c-Kit+) cell subpopulations in 14.5 dpc mel-18−/− or littermate control FT. Total RNA was isolated from indicated cells using TRIzol reagent (Invitrogen Life Technologies) and was reverse transcribed using the SuperScript II RT-PCR system (Invitrogen Life Technologies) with oligo(dT) primer or random hexamer primer. Real-time PCR was performed using the ABI PRISM 7000 Sequence Detection System (Applied Biosystems) according to the manufacturer’s instructions. The 50-μl amplification reaction mixture contained 5 μl of cDNA (1/10 dilution), 25 μl of TaqMan universal PCR master mix (Applied Biosystems), and 200 nM each of primers. Each sample was amplified in triplicate using primers and probes specific for mel-18. A two-step cycling protocol with combined primer annealing and elongation at 60°C was used. The TaqMan ribosomal RNA control reagent (Applied Biosystems) was used as the endogenous normalization standard. For semiquantitative PCR, cDNAs were amplified using various oligonucleotide primers chosen on the basis of Oligo 4.0 software (National Biosciences). The primers used for the PCR were as follows: mel-18 (real-time PCR), 5′-GCGACGGGACTTCTATGCA-3′ and 5′-CGCCTTCGTAGAACTCAATGG-3′; mel-18, 5′-CGGAGAATGGAGATGGGGACAAGGAGAAG-3′ and 5′-AAGGTGGAGTGGGGGAAGTAGGATGGGTAG-3′; G3PDH, 5′-GTGAAGGTCGGTGTGAACGGAT-3′ and 5′-CAGAAGGGGCGGAGATGATGAC-3′; Notch-1, 5′-CCCAGCAGGTGCAGCCACAG-3′ and 5′-GGTGATCTGGGACGGCATGG-3′; Hes-1, 5′-GCCAGTGTCAACACGACACCGG-3′ and 5′-TCACCTCGTTCATGCACTCG-3′; GATA-3, 5′-TCGGCCATTCGTACATGGAA-3′ and 5′-GAGAGCCGTGGTGGATGGAC-3′; LEF-1, 5′-AACTCTGCGCCACCGATGAG-3′ and 5′-AGAAAAGTGCTCGTCGCTGT-3′; TCF-1, 5′-GCCAGCCTCCACATGGTGTC-3′ and 5′-CGCGTGAGGGATGGCTGCTG-3′; Deltex-1, 5′-CCACTGCTACCTACCCAACAAT-3′ and 5′-AGGCTAGAGGCAAGGCAAAAGG-3′; RBP-J, 5′-CACAGACAAGGCAGAATACACG-3′ and 5′-TGTAGGTGAAGGTAAGGCTGGT-3′; and E2A, 5′-CATCCATGTCCTGCGAAGCCAC-3′ and 5′-TTCTTGTCCTCTTCGGCGTCGG-3′. PCR amplification was performed using the GeneAmp PCR system 9700 (Applied Biosystems). The cycle number was 24–34, because the amplification was found to be within the linear range (data not shown).

PCR product was electrophoresed on 2% agarose gel, blotted onto a nylon membrane, and analyzed by Southern blot hybridization with 33P-labeled probes. The signals were processed and quantified by an imaging analyzer (BAS 2000; Fuji Photo Film). The following probes were used: mel-18, 5′-ACCTGGCAAAGTTCCTCCGC-3′; G3PDH, 5′-GGTGGACCTGACCTGCCGTCTAGAAAAAC-3′; Notch-1, 5′-ACCCCTTCCTCACCCCATCCCC-3′; Hes-1, 5′-CCGCCGCGCTCAGCACAGACCC-3′; GATA-3, 5′-GGTATGCCGCCCGCCTCTGCTG-3′; LEF-1, 5′-GATGCCCAATATGAACAGCGACCC-3′; TCF-1, 5′-ACCCCCTGTCCCCTTCCTGCGG-3′; Deltex-1, 5′-TGCTCATCACCGCCTGGGAACG-3′; RBP-J, 5′-GTCCGCCAGCCAGTCCAGGTTC-3′; and E2A, 5′-ATGCCAGCCTCCCCAGCCAGCC-3′.

Embryos (12.5 dpc) were embedded in OCT compound (Miles) and snap-frozen. Freshly cut 5-μm sections were fixed with acetone at room temperature for 2 min. Sections were incubated in rabbit anti-keratin Ab (wide spectrum screening; DakoCytomation) or rabbit anti-Ikaros Ab (54), then incubated in HRP-conjugated goat anti-rabbit IgG (Vector Laboratories) at room temperature. Peroxidase activity was developed with 0.1% 3,3′-diaminobenzidine and 0.02% H2O2 in PBS. Sections were then counterstained with hematoxylin.

Freshly cut 5-μm sections were fixed with acetone at room temperature for 2 min. Sections were incubated in rabbit anti-keratin Ab for 2 h, then incubated in goat anti-rabbit IgG-Texas Red conjugate (Molecular Probes) for 1 h at room temperature. TUNEL reaction was performed on sections using an In Situ Cell Death Detection kit, Fluorescein (Roche), according to the manufacturer’s protocol. Briefly, sections were fixed in 4% paraformaldehyde in PBS for 10 min at room temperature, incubated in 0.1% Triton X-100/0.1% sodium citrate for 2 min at 4°C, then incubated in TUNEL reaction mixture for 1 h at 37°C.

The amount of nuclear DNA was determined by PI staining as follows. DN1 and DN2 14.5E FT cells were fixed in 50% ethanol at 4°C for 30 min and incubated in PBS containing 1 mg/ml RNase (Sigma-Aldrich) at 37°C for 20 min. The cells were washed in PBS, resuspended in PBS containing 100 μg/ml PI, and analyzed by a FACSCalibur (BD Biosciences).

The in vitro survival assay was conducted as previously described (35). Sorted cells were cultured in 96-well, flat-bottom plates in a volume of 0.15 ml of RPMI 1640 with 10% FCS and 2-ME. SCF (100 ng/ml) and IL-7 (150 U/ml) were added to the medium. After 0 or 20 h, cells were harvested, and the numbers of surviving T cell precursors were counted by flow cytometry.

Tst-4, a thymus-derived fibroblast cell line (36), was transfected with retrovirus containing Delta-like-1 (DL1) gene. DL1 expressing TSt-4 efficiently support the differentiation of T cells from Linc-Kit+ progenitors of FL and bone marrow in a monolayer culture, whereas they suppress B cell generation (H. Kawamoto, unpublished observation). IL-7R+ cells in 12.5 dpc mel-18−/− or control FL were sorted and cocultured with Tst-4/DL1 cell in FTOC medium. After day 3 or 6 of culture, Ly5+/Thy1.2+ cells were sorted.

Dicer siRNA for mel-18 or GFP were generated using the Dicer siRNA Generation Kit (Gene Therapy Systems). In brief, cDNA construct for mel-18 with T7 promoters at the 717 and 3′ ends were made by PCR and subjected to in vitro transcription to produce dsRNA. Dicer siRNAs were produced by digestion of dsRNA with recombinant Dicer enzyme. KKC cells (1 × 105) were transfected with 125 ng of dicer-siRNA for either mel-18 or GFP using Lipofectamine 2000 (Invitrogen Life Technologies). After 48 h, cells were harvested and total RNA was extracted and subjected to semiquantitative RT-PCR analysis.

It has been previously reported that the paucity of cell expansion and maturation arrest at the DN stage were observed in adult thymocytes from mel-18−/− mice on a mixed C57BL/6(B6)×129 background (14). Subsequently, our group and others noticed that the CD4/CD8 profile of thymocytes in mel-18−/− mice became comparable to that of mel-18+/+ mice after an extensive backcross to B6 mice (>10 times; Fig. 1,A) (18). Hereafter, experiments were performed exclusively with mice on a B6 background. In these 6-wk-old male B6 mel-18−/− mice, a severe reduction in total thymocyte number was observed despite the relatively normal CD4/CD8 profile (Fig. 1, A and B). The total thymocyte number in mel-18−/− mice was reduced to <10% of the wild-type number, even in 2-wk-old mice, and was further decreased to ∼2% of the wild-type number in 10-wk-old mice (Fig. 1,B). Next, we investigated the cellularity of the immature, Lin thymocyte population subdivided by CD44/CD25 (DN1-DN4) expression criteria. The severe reduction of the absolute cell number (to <10% of normal) was apparent at the most immature stage, DN1. No maturation arrest among DN subpopulations was observed (Fig. 1, A and B). From these results, we speculated that the severe reduction of the cell number might begin at the DN1 stage or earlier in mel-18−/− mice.

FIGURE 1.

Phenotypic characterization of mel-18−/− thymocytes and expression level of mel-18 during T cell development. A, A representative flow cytometric analysis of CD4 vs CD8 on total thymocytes, and analysis of CD44 vs CD25 on Lin DN thymocytes from 6-wk-old male wild-type control littermate (mel-18+/+) and mel-18−/− mice. Percentages of CD4 SP, CD8 SP, DP, DN, and DN1–4 cells are indicated. B, Absolute cell numbers were calculated for total thymocytes and the thymocyte subpopulations described in A. The bars represent the mean ± SD. The cell number of total thymocytes from adult mel-18+/+, mel-18+/−, and mel-18−/− mice, 2–10 wk old, are plotted. C, Total thymocytes from 6-wk-old mel-18+/+ and mel-18−/− mice were stained for lineage markers, CD44, CD25, c-Kit, and IL-7Rα, and 5 × 106 cells were analyzed. The percentage of total thymocytes and the absolute number of cells within the indicated region are shown. D, Bone marrow cells from femurs and tibiae were stained for lineage markers, Sca-1, c-Kit, and IL-7Rα. A representative flow cytometric analysis of c-Kit vs IL-7Rα on gated LinSca-1+ bone marrow cells is shown. Total cell numbers of KSL and IL-7R+ from femurs and tibiae were calculated and presented in graphic histograms. The bars represent the mean ± SD. E, Real-time PCR analysis of mel-18 expression in thymocyte and bone marrow cell fractions described in A and D from adult male B6. Expression in DP was defined as 1.

FIGURE 1.

Phenotypic characterization of mel-18−/− thymocytes and expression level of mel-18 during T cell development. A, A representative flow cytometric analysis of CD4 vs CD8 on total thymocytes, and analysis of CD44 vs CD25 on Lin DN thymocytes from 6-wk-old male wild-type control littermate (mel-18+/+) and mel-18−/− mice. Percentages of CD4 SP, CD8 SP, DP, DN, and DN1–4 cells are indicated. B, Absolute cell numbers were calculated for total thymocytes and the thymocyte subpopulations described in A. The bars represent the mean ± SD. The cell number of total thymocytes from adult mel-18+/+, mel-18+/−, and mel-18−/− mice, 2–10 wk old, are plotted. C, Total thymocytes from 6-wk-old mel-18+/+ and mel-18−/− mice were stained for lineage markers, CD44, CD25, c-Kit, and IL-7Rα, and 5 × 106 cells were analyzed. The percentage of total thymocytes and the absolute number of cells within the indicated region are shown. D, Bone marrow cells from femurs and tibiae were stained for lineage markers, Sca-1, c-Kit, and IL-7Rα. A representative flow cytometric analysis of c-Kit vs IL-7Rα on gated LinSca-1+ bone marrow cells is shown. Total cell numbers of KSL and IL-7R+ from femurs and tibiae were calculated and presented in graphic histograms. The bars represent the mean ± SD. E, Real-time PCR analysis of mel-18 expression in thymocyte and bone marrow cell fractions described in A and D from adult male B6. Expression in DP was defined as 1.

Close modal

Recent reports have revealed that the DN1 stage comprises at least two subpopulations, c-KithighIL-7Rαneg/low (c-Kithigh) and c-Kitneg/low IL-7Rα+, and that only cells within the c-Kithigh pool have T-lineage potential in adult mice (37). Additionally, our preliminary observations suggested that c-Kithigh DN1, but not c-Kitneg/low DN1, cells retained T-lineage reconstitution potential (data not shown), and a recent study supported our data (38). Therefore, we examined whether these c-Kithigh cells were affected by a lack of mel-18. The absolute cell number of c-Kithigh cells in mel-18−/− mice was reduced to 3% of that observed in mel-18+/+ mice (1,200 cells and 40,600 cells/thymus, respectively; Fig. 1 C).

Next, we investigated whether these cellular defects occurred after migration into the thymus or at a prethymic progenitor stage in adult bone marrow. We analyzed the absolute number of hemopoietic progenitors, defined as KSL cells, and so-called common lymphoid progenitors, defined as LinSca-1+c-KitlowIL-7Rα+ (7R+) cells (39) in bone marrow. There was no significant difference between the numbers of 7R+ and KSL cells in mel-18+/+ and mel-18−/− mice (Fig. 1 D).

The expression level of mel-18 was assessed in normal cell populations by real-time PCR (Fig. 1 E). The level of mel-18 mRNA was higher in the DN subset than in the DP and SP subsets. Within the DN population, the mel-18 expression level was highest in DN1 and was gradually decreased in the stages leading up to the more differentiated DN4. The level in DN1 was higher than that in KSL and 7R+ cells in bone marrow. The level of mel-18 mRNA in the c-Kithigh subset was comparable to that in the entire population of DN1 cells (data not shown). Therefore, these data show that the mel-18 gene is expressed at a higher level in DN1 thymocytes, including c-Kithigh cells, than any at other stage during the developmental process from HSC to mature T cells in adult mice. Based on our data, we speculate that the thymic atrophy in adult mel-18−/− mice is probably caused by defects in the earliest c-Kithigh DN1 cells, although it remains to be clarified whether migration into the thymus is affected in adult mel-18−/− mice.

We also examined whether early T cell development was impaired in the mel-18−/− fetus. In the FT at 12.5 dpc, when thymocytes are exclusively composed of DN1 cells (Fig. 2,B), we found no significant difference in the total number of thymocytes between mel-18+/− and mel-18−/− mice; however, at 13.5 dpc, the number of thymocytes in the mel-18−/− fetus was reduced to about half; furthermore, from 14.5 dpc to birth, the number of thymocytes was reduced to less than one-third of the number in the mel-18+/− fetus (Fig. 2,A). However, we observed a small decrease in the percentage of DN2 cells, but no differentiation arrest in CD44/25 profile of the FT at 12.5–14.5 dpc (Fig. 2 B). These observations suggested that a lack of mel-18 also affected the development of early T progenitors in the fetus.

FIGURE 2.

Analysis of fetal thymocytes and prethymic lymphoid progenitors in mel-18−/− mice. A, Cell number of total thymocytes from a control littermate (mel-18+/−) and a mel-18−/− fetus. The cell numbers of total thymocytes from 12.5 dpc to neonatal mice are plotted. There were no differences in cell number or FACS profiles between mel-18+/+ and mel-18+/− fetuses in this period (data not shown). B, Representative flow cytometric analysis of CD44 vs CD25 on fetal thymocytes from mel-18+/− and mel-18−/− fetus in 12.5, 13.5, and 14.5 dpc fetuses. The percentages of DN1-DN4 cells are indicated. C, Fetal blood cells in 13.5 dpc fetuses were stained for lineage markers (CD3, B220, CD19, Gr-1, Ter119, Thy1.2, and NK1.1), c-Kit, and IL-7Rα. Representative flow cytometric analyses are shown. Two experiments with independent litters were performed with similar results. D, Distributions of epithelial and hemopoietic progenitor cells in the thymus anlage. Serial frozen sections of 12.5 dpc mel-18+/− and mel-18−/− embryos were incubated with anti-keratin or anti-IKAROS Abs. The lines indicate the border between the epithelial region and the mesenchymal layer of the thymus anlage. Scale bar, 50 μm.

FIGURE 2.

Analysis of fetal thymocytes and prethymic lymphoid progenitors in mel-18−/− mice. A, Cell number of total thymocytes from a control littermate (mel-18+/−) and a mel-18−/− fetus. The cell numbers of total thymocytes from 12.5 dpc to neonatal mice are plotted. There were no differences in cell number or FACS profiles between mel-18+/+ and mel-18+/− fetuses in this period (data not shown). B, Representative flow cytometric analysis of CD44 vs CD25 on fetal thymocytes from mel-18+/− and mel-18−/− fetus in 12.5, 13.5, and 14.5 dpc fetuses. The percentages of DN1-DN4 cells are indicated. C, Fetal blood cells in 13.5 dpc fetuses were stained for lineage markers (CD3, B220, CD19, Gr-1, Ter119, Thy1.2, and NK1.1), c-Kit, and IL-7Rα. Representative flow cytometric analyses are shown. Two experiments with independent litters were performed with similar results. D, Distributions of epithelial and hemopoietic progenitor cells in the thymus anlage. Serial frozen sections of 12.5 dpc mel-18+/− and mel-18−/− embryos were incubated with anti-keratin or anti-IKAROS Abs. The lines indicate the border between the epithelial region and the mesenchymal layer of the thymus anlage. Scale bar, 50 μm.

Close modal

To confirm this, we had to rule out the following possibilities: first, a decreased number of prethymic progenitors in FL or FB, and second, an obstructed migration of these progenitors into the FT. We performed flow cytometric analysis to detect the number of Linc-Kithigh cells (hemopoietic progenitors) and Linc-KithighIL-7R+ (IL-7R+) cells in FL and FB. We have previously shown that IL-7R+ cells in both 12.5 dpc FL and FB contained mostly T cell lineage-restricted progenitors, and those cells isolated from 13.5 dpc FB also represented T cell progenitors (40, 41). In 12.5 dpc FL and FB, the percentage of IL-7R+ cells in mel-18−/− fetuses was the much same as that in mel-18+/− fetuses (data not shown). Furthermore, at 13.5 dpc, in which the cell number in mel-18−/− FT was decreased to about half that in mel-18+/− FT, there was no significant difference in the percentage of IL-7R+ cells in FB between the two genotypes (Fig. 2,C). To rule out the second possibility, the impairment of migration, the distribution of hemopoietic progenitors and thymic epithelial cells was investigated in sagittal sections of whole embryos at 12.5 dpc by immunohistochemistry. Hemopoietic progenitors and thymic epithelial cells were determined by staining with an anti-IKAROS or an anti-keratin Ab, respectively. Hemopoietic progenitors normally begin immigration into the thymic mesenchymal layer at 11.5 dpc and enter the epithelial region at 12.5 dpc (42). IKAROS-positive cells that reside in the mesenchymal layer surrounding the epithelial region were regarded as immigrating progenitors. Taken together with the data presented in Fig. 2, A, C, and D, the number of immigrating progenitors in mel-18−/− mice was comparable to that in mel-18+/− mice (Fig. 2 D). These results indicate that the mel-18 deficiency affected neither prethymic T progenitors nor their migration into the thymus anlage.

The above results raised the question of whether the reduced cell number of mel-18−/− FT was caused by impaired expansion of early thymocytes or an inability on the part of the thymic environment to support thymocyte development. To assess the potential of the thymic lobe to support T cell development, we performed a FTOC. DN1 cells (CD44+CD25c-Kithigh; Ly5.1/5.2 F1) were transferred into each lobe of mel-18+/− or mel-18−/− (Ly5.2) FTs that had been treated with dGuo. After a 10-day culture period, the number of donor-derived thymocytes, Ly5.1-positive cells, was counted and analyzed. No significant difference was observed in cell number or in the CD4/CD8 expression profile of the recovered Ly5.1+ cells (data not shown), indicating that the mel-18−/− thymic lobe sustained the capacity to support immature thymocyte development. Next, 14.5 dpc FT lobes from either mel-18+/− or mel-18−/− mice were cultured for 3 and 6 days. As shown in Fig. 3 A, we found that the early expansion of immature thymocytes was impaired in mel-18−/− FT, specifically on day 3, whereas there appeared to be normal maturation of T cell development, as defined by the CD4/CD8 profiles (data not shown).

FIGURE 3.

Impaired expansion of fetal DN1 cells in mel-18−/− mice. A, A total of 14.5 dpc FT lobes from either mel-18+/− or mel-18−/− fetus was cultured for 3 and 6 days in organ. Indicated are viable cell numbers per lobe. B, Single CD44+CD25c-Kit+ (DN1) FT cells from 14.5 dpc mel-18+/− or mel-18−/− fetus (Ly5.2) were picked up under microscopic visualization and were seeded into wells containing a dGuo-treated FT lobe (Ly5.1/5.2 F1) in the presence of IL-7 and SCF. After 14 days of culture under HOS conditions, cells in each well were analyzed by flow cytometry. Ly5.1-positive cells were gated out (data not shown). The numbers of T progenitors per 14.5 dpc FT CD44+CD25c-Kit+ cells are scored among 30 cells (reconstitution frequency). The numbers of cell recoveries in reconstituted clones are plotted (single DN1 cell proliferation). Bars denote the mean value (mean, 12,073 and 6,730, respectively). C, Absolute cell numbers were calculated for total thymocytes from Rag1−/−mel-18+/+, Rag1−/−mel-18+/−, or Rag1−/−mel-18−/− fetus at 17.5 dpc. Shown is a 17.5 dpc fetal thymus from Rag1−/−mel-18+/+ and Rag1−/−mel-18−/− fetus. Scale bar, 1 mm.

FIGURE 3.

Impaired expansion of fetal DN1 cells in mel-18−/− mice. A, A total of 14.5 dpc FT lobes from either mel-18+/− or mel-18−/− fetus was cultured for 3 and 6 days in organ. Indicated are viable cell numbers per lobe. B, Single CD44+CD25c-Kit+ (DN1) FT cells from 14.5 dpc mel-18+/− or mel-18−/− fetus (Ly5.2) were picked up under microscopic visualization and were seeded into wells containing a dGuo-treated FT lobe (Ly5.1/5.2 F1) in the presence of IL-7 and SCF. After 14 days of culture under HOS conditions, cells in each well were analyzed by flow cytometry. Ly5.1-positive cells were gated out (data not shown). The numbers of T progenitors per 14.5 dpc FT CD44+CD25c-Kit+ cells are scored among 30 cells (reconstitution frequency). The numbers of cell recoveries in reconstituted clones are plotted (single DN1 cell proliferation). Bars denote the mean value (mean, 12,073 and 6,730, respectively). C, Absolute cell numbers were calculated for total thymocytes from Rag1−/−mel-18+/+, Rag1−/−mel-18+/−, or Rag1−/−mel-18−/− fetus at 17.5 dpc. Shown is a 17.5 dpc fetal thymus from Rag1−/−mel-18+/+ and Rag1−/−mel-18−/− fetus. Scale bar, 1 mm.

Close modal

To acquire a more detailed understanding of the impaired expansion, we examined the T cell reconstitution frequency of DN1 cells at the single-cell level using the progenitor assay. DN1 cells (Ly5.2) from 14.5 dpc mel-18+/− or mel-18−/− FT were cultured with a dGuo-treated FT lobe of B6 mice (Ly5.1/5.2 F1) at one cell per lobe under HOS conditions. After 14 days, the thymocytes generated in the culture were collected and analyzed by flow cytometry to determine the potential of the initial progenitor cell for generating T cells. The frequency of reconstituted T progenitors from mel-18−/− DN1 cells was reduced to 8 of 30 (26.7%), in contrast to the 17 of 30 (56.7%) observed for mel-18+/−. Furthermore, the number of cells generated from a single mel-18−/− DN1 cell was about half that generated by a mel-18+/− DN1 cell (Fig. 3,B). These results suggest that the reduced cell number of fetal thymocytes was due to a decreased number of early T progenitors and the impaired expansion of the T progenitors in mel-18−/− mice. Taken together with the results shown in Fig. 2 A, in which thymocyte number in mel-18−/− FT was obviously reduced at 14.5 dpc, we concluded that mel-18 deficiency impairs the pre-β proliferation of early T progenitors.

To confirm this, we established breeding mice, which lacked both the Rag1 and mel-18 genes (Rag1−/−mel-18−/−). Rag1 is essential for TCR gene rearrangement, and thymocytes from Rag1 mutant mice arrest at the DN3 stage. Therefore, backcrossing to Rag1 mutant mice makes it possible to investigate pre-β proliferation specifically. The number of thymocytes in the Rag1−/−mel-18−/− fetus was reduced to less than one-third that observed in either the Rag1−/−mel-18+/+ or the Rag1−/−mel-18+/− fetus at 17.5 dpc (Fig. 3 C). These results clearly demonstrate that mel-18 is required for the expansion of early T progenitors before TCR β-chain gene rearrangement.

It was previously reported that the defect in T cell development in mel-18−/− mice was due to the impaired proliferation of thymocytes in response to IL-7, although IL-7 and IL-7R expression and the JAK/STAT signaling pathway were found to be normal (14). These studies, however, were primarily focused on the total thymocyte population and not specifically on early T progenitors. We chose to investigate the cell cycle parameters of T progenitors in mel-18+/− and mel-18−/− FT in two different ways. First, DN1/DN2 cells from 13.5 dpc FT were stained with PI, and cell cycle profiles were analyzed by flow cytometry. The proportion of cells in S-G2-M phase in mel-18−/− DN1 cells was comparable to that in mel-18+/− DN1 cells (34.9 ± 3.4 and 35.3 ± 0.4%, respectively). Similar results were observed for DN2 cells (Fig. 4 A). Second, to assess the number of cells containing newly synthesized DNA, 14.5 dpc fetuses were pulse-labeled with BrdU in vivo. DN1/DN2 cells were sorted and double-stained with an anti-BrdU Ab and Hoechst 33342. Consistent with the cell cycle profiles, we could not find any difference in the proportion of BrdU-positive cells between mel-18+/− and mel-18−/− DN1/DN2 cells (both ∼40%; data not shown). Therefore, we considered that the reduction in fetal thymocytes did not correlate with a change in mitotic activity in mel-18−/− mice.

FIGURE 4.

Dysfunction of DN1 cells in mel-18−/− fetus is possibly caused by increased susceptibility to cell death. A, Cell cycle analysis of sorted DN1(CD44+CD25c-Kit+) and DN2(CD44+CD25+c-Kit+) cells from 14.5 dpc mel-18+/− or mel-18−/− fetus. Three independent experiments were performed with similar results. B, Serial frozen sections of 12.5 dpc mel-18+/− or mel-18−/− fetus were stained with anti-keratin (red) and subsequently processed for TUNEL assay (green). The lines indicate the border between the epithelial region and the mesenchymal layer of the thymus anlagen. The numbers of TUNEL-positive cells within the epithelial region were counted on every two serial sections (5-μm thick) throughout the anlagen. The total numbers of TUNEL-positive cells from the sections of a single anlage of mel-18−/− mice were almost twice those of control littermates (mel-18+/−; total, 155 and 84, respectively). The representative results of two experiments (two mel-18+/− and three mel-18−/− fetuses) are shown. Scale bar, 50 μm. C, In vitro survival assay. Sorted DN1 and DN2 cells FT cells from a 14.5 dpc mel-18+/− or mel-18−/− fetus were cultured with IL-7 or SCF. After 0 or 20 h, the numbers of viable cells were counted by flow cytometry. Three independent experiments were performed with similar results.

FIGURE 4.

Dysfunction of DN1 cells in mel-18−/− fetus is possibly caused by increased susceptibility to cell death. A, Cell cycle analysis of sorted DN1(CD44+CD25c-Kit+) and DN2(CD44+CD25+c-Kit+) cells from 14.5 dpc mel-18+/− or mel-18−/− fetus. Three independent experiments were performed with similar results. B, Serial frozen sections of 12.5 dpc mel-18+/− or mel-18−/− fetus were stained with anti-keratin (red) and subsequently processed for TUNEL assay (green). The lines indicate the border between the epithelial region and the mesenchymal layer of the thymus anlagen. The numbers of TUNEL-positive cells within the epithelial region were counted on every two serial sections (5-μm thick) throughout the anlagen. The total numbers of TUNEL-positive cells from the sections of a single anlage of mel-18−/− mice were almost twice those of control littermates (mel-18+/−; total, 155 and 84, respectively). The representative results of two experiments (two mel-18+/− and three mel-18−/− fetuses) are shown. Scale bar, 50 μm. C, In vitro survival assay. Sorted DN1 and DN2 cells FT cells from a 14.5 dpc mel-18+/− or mel-18−/− fetus were cultured with IL-7 or SCF. After 0 or 20 h, the numbers of viable cells were counted by flow cytometry. Three independent experiments were performed with similar results.

Close modal

Consequently, to explain the impaired expansion of early T progenitors in mel-18−/− FT, we performed TUNEL staining to detect apoptotic cells in 12.5 dpc FT. As shown in Fig. 4,B, we observed a 2-fold increase in the number of TUNEL-positive cells in mel-18−/− FT compared with that in mel-18+/− FT (see Fig. 4,B). This increase in the number of TUNEL-positive cells could explain the data shown in Fig. 3 B; that is, early T progenitors in mel-18−/− FT were prone to cell death during the process of pre-β proliferation, and this resulted in a low frequency of T progenitors within the DN1 cell population and impaired expansion from a single DN1 cell. We suggest that the reduction of fetal thymocytes is due to an increased susceptibility to cell death or an insufficient presence of survival signals in early T progenitors in mel-18−/− mice.

Because IL-7R signaling was known to be required for cell survival during immature T cell development, especially from the DN1 to the DN3 stage (43, 44), we next examined the possibility that the increased cell death in mel-18−/− FT was due to an impairment of IL-7R signaling. In an in vitro survival assay, DN1/2 cells in mel-18−/− FT displayed a normal response to SCF and IL-7, suggesting that the increased cell death of T progenitors in mel-18−/− FT is not correlated with IL-7 or SCF signaling (Fig. 4 C).

To identify underlying mechanisms involved in this dysfunction of early T progenitors in mel-18−/−] FT, we analyzed the expression of several genes that are involved in early thymocyte development. As shown in Fig. 5, mel-18 deficiency had no effect on the expression of many transcription factors, including GATA-3, TCF-1, LEF-1, and E2A.

FIGURE 5.

The expression of Hes-1, a target gene of Notch-1 signal, is decreased in mel-18−/− DN1 and DN2 cells. Semiquantitative RT-PCR was performed with primers specific for each gene on cDNAs obtained from sorted DN1 and DN2 cells in 14.5 dpc mel-18−/− or littermate control FT. After RT-PCR, the products were subjected to Southern blot analysis. As an internal control, the expression of G3PDH was investigated. The results were representative of two independent litters.

FIGURE 5.

The expression of Hes-1, a target gene of Notch-1 signal, is decreased in mel-18−/− DN1 and DN2 cells. Semiquantitative RT-PCR was performed with primers specific for each gene on cDNAs obtained from sorted DN1 and DN2 cells in 14.5 dpc mel-18−/− or littermate control FT. After RT-PCR, the products were subjected to Southern blot analysis. As an internal control, the expression of G3PDH was investigated. The results were representative of two independent litters.

Close modal

Surprisingly, the expression of the Hes-1 gene, one of the target genes of the Notch-1 signal, was remarkably decreased in mel-18−/− DN1 and DN2 cells compared with mel-18+/− cells. In contrast, there were no significant differences in the expression of Notch-1, RBP-J, or Deltex-1. These results suggested that mel-18 might be involved in Notch-1 signaling, especially in the regulation of Hes-1 expression, in early T progenitors. Moreover, we investigated the expression of Hes-1 in DN1/DN2 cells in adult mel-18−/− mice, and indeed, we observed down-regulation of the Hes-1 gene similar to that observed in FT (data not shown).

To confirm that mel-18 is involved in Hes-1 expression induced by Notch-1 signal, we set up an in vitro coculture system for studying T cell development with DL1-expressing stroma cell. Recently, it has been reported that OP-9 cells ectopically expressing DL1 acquired a capacity for T cell differentiation in monolayer culture (31). Therefore, we used a similar system, with a thymus-derived fibroblast cell line, TSt-4, transfected with the DL1 gene. IL-7R+ cells (Linc-Kit+Il-7R+) in 14.5 dpc FL were cultured on TSt-4 stroma cells expressing DL1 (TSt-4/DL1). We found that TSt-4/DL1 cells suppressed B cell generation while inducing T cell differentiation (data not shown). When IL-7R+ cells from 12.5 dpc mel-18+/− or mel-18−/− FL were sorted and cocultured with TSt-4/DL1 cells for 5 days, the number of cells generated from IL-7R+ cells in mel-18−/− FL was less than one-third that in mel-18+/− cells (data not shown). Using this system, we examined the precise time course of Hes-1 expression. After 3 and 6 days of culture, T precursors (Thy1.2+ cells) were sorted, and the expression of the Hes-1 gene was analyzed by semiquantitative RT-PCR Southern blot analysis. Indeed, on day 6, the expression of Hes-1 in mel-18−/− Thy1.2 cells was decreased by 20% compared with the expression in mel-18+/− cells, whereas Hes-1 expression in mel-18−/− cells on day 3 was comparable to that in the control (Fig. 6,A). From these results, we advanced the hypothesis that mel-18 was engaged in maintaining the active transcriptional state of the Hes-1 gene once induced by DL1. To investigate this hypothesis, we first surveyed the expression of Hes-1 and mel-18 in several DN thymocyte cell lines. KKC cells, a DN1 thymocyte cell line (45), displayed Hes-1 expression in the absence of the Notch-1 ligand, and the mel-18 gene was also highly expressed in this cell line (data not shown). Because the Hes-1 gene seemed to be maintained in an active state in KKC cells, we investigated, using this cell line, whether the reduction of mel-18 expression affected the expression level of the Hes-1 gene. KKC cells were transfected with a dicer-siRNA for either mel-18 or GFP. The expression of mel-18 was reduced to ∼2% by the siRNA for mel-18 compared with the expression level observed upon application of the siRNA for GFP (Fig. 6 B). Under these conditions, the expression level of the Hes-1 gene was decreased to 5% by the siRNA for mel-18. Based on these results, we propose that mel-18 is indispensable for maintaining the active state of the Hes-1 gene once it is established by the Notch-1 signal.

FIGURE 6.

Mel-18 is associated with maintenance of Hes-1 gene expression. A, Sorted IL-7R+ cells in 12.5 dpc mel-18+/− or mel-18−/− FL were cocultured with TSt-4/DL1 for 3 and 6 days. Expression of the Hes-1 gene was assessed by semiquantitative RT-PCR Southern blot analysis using cDNAs isolated from sorted T precursor cells (Ly5+Thy1.2+). Arbitrary densitometric ratios of Hes-1/G3PDH are shown. The expression in mel-18+/− cells was defined as 1.0 in each panel. B, KKC cells express Hes-1, Notch-1, and mel-18 mRNA in the absence of Notch-1 ligands (data not shown). KKC cells were transfected with dicer-siRNA for either mel-18 or GFP. mRNA levels of mel-18, Hes-1, and G3PDH were determined by semiquantitative RT-PCR and Southern blot analysis with 4-fold serial dilution of template cDNA. The data with asterisks correspond to each other. The relative ratio of mel-18/G3PDH was calculated with cells transfected with mel-18 siRNA or GFP siRNA. The expression level of mel-18 was decreased to 0.02 by mel-18 siRNA compared with that in siRNA for GFP. Arbitrary densitometric ratios of Hes-1/G3PDH are indicated. The ratio in siRNA for GFP was defined as 1.0.

FIGURE 6.

Mel-18 is associated with maintenance of Hes-1 gene expression. A, Sorted IL-7R+ cells in 12.5 dpc mel-18+/− or mel-18−/− FL were cocultured with TSt-4/DL1 for 3 and 6 days. Expression of the Hes-1 gene was assessed by semiquantitative RT-PCR Southern blot analysis using cDNAs isolated from sorted T precursor cells (Ly5+Thy1.2+). Arbitrary densitometric ratios of Hes-1/G3PDH are shown. The expression in mel-18+/− cells was defined as 1.0 in each panel. B, KKC cells express Hes-1, Notch-1, and mel-18 mRNA in the absence of Notch-1 ligands (data not shown). KKC cells were transfected with dicer-siRNA for either mel-18 or GFP. mRNA levels of mel-18, Hes-1, and G3PDH were determined by semiquantitative RT-PCR and Southern blot analysis with 4-fold serial dilution of template cDNA. The data with asterisks correspond to each other. The relative ratio of mel-18/G3PDH was calculated with cells transfected with mel-18 siRNA or GFP siRNA. The expression level of mel-18 was decreased to 0.02 by mel-18 siRNA compared with that in siRNA for GFP. Arbitrary densitometric ratios of Hes-1/G3PDH are indicated. The ratio in siRNA for GFP was defined as 1.0.

Close modal

In Fig. 7, we provide a model explaining how mel-18 regulates the expansion of early T progenitors before ΤCRβ rearrangement by maintaining Hes-1 gene expression.

FIGURE 7.

Schematic illustration of mel-18 function in the pre-β rearrangement proliferation of early T progenitors through the maintenance of Hes-1 gene expression.

FIGURE 7.

Schematic illustration of mel-18 function in the pre-β rearrangement proliferation of early T progenitors through the maintenance of Hes-1 gene expression.

Close modal

It has been previously shown that expression levels of individual PcG genes varied during hemopoiesis (10, 46). Mutant mice lacking each constituent of a PRC1-like complex tended to display a characteristic defect at the stage of hemopoiesis in which the particular gene under investigation was highly expressed. For instance, the expression of mel-18 is high in immature DN thymocytes, especially in DN1 cells, and the lack of mel-18 affects DN1 cells in the adult thymus (Fig. 1). In addition, Bmi-1 and Rae-28 are essential for the self-renewal activity of adult or fetal HSCs, respectively, in which the expression of each gene is particularly high (46, 47, 48). These observations suggest that the subunit stoichiometry of the PRC1-like complex changes during hemopoiesis, and that a stage-specific PRC1-like complex can determine a distinct spectrum of target genes. Consequently, this complex plays a specific role at each developmental stage. We believe that the mel-18-dominant PRC-1-like complex (PcG/mel-18) may be responsible for the expansion of early T progenitors.

In T cell development, the expansion of T progenitors that have migrated into the thymus is required for clonal diversification of TCR, and this diversification profoundly contributes to the recognition of a wide variety of Ags. It has been proposed that there are two independent expansion phases of T progenitors. The so-called post-β proliferation takes place immediately after TCRβ rearrangement, which makes it possible to generate diverse TCR α-chains from a single TCR β-chain (21, 22). In contrast, the pre-β proliferation, which is predicted to be on the order of 1000-fold, starts from DN1 cells, and this proliferation is also essential for the formation of diversified TCR β-chains (20, 41). The Rag1-mel-18 double-knockout (Fig. 3 C) provides clear evidence that the lack of mel-18 impairs the pre-β proliferation of early T progenitors in FT, indicating that the PcG/mel-18 complex plays a role in developing the diversified immune system by supporting the pre-β proliferation of early T progenitors.

We observed that expression of the Hes-1 gene was dramatically decreased in mel-18−/− DN1 and DN2 cells, although other Notch-1 signal-related genes were expressed normally (Fig. 5). Hes-1 is a basic helix-loop-helix transcriptional factor and functions as a negative or positive regulator of cell growth and differentiation in various systems (49). In T cell development, prethymic T cell progenitors in FL express a relatively low level of Hes-1, even though they express Notch-1 receptor. After migration into the thymus, T progenitors interact with thymic stromal cells and receive a signal through the Notch-1 receptor, and the Hes-1 gene is up-regulated by this signal (50) (Fig. 6,B). We note that a mel-18 deficiency affects early T progenitors after migration into the thymus, not at the prethymic stage (Figs. 1 and 2, B and C), demonstrating that the consequences of mel-18 deficiency correlate with the expression pattern of the Hes-1 gene. Interestingly, the phenotype of fetal thymocytes from Hes-1−/− mice is strikingly similar to that of mel-18−/− mice (Figs. 2 and 3). Prethymic T progenitors in FL and FB were not affected in Hes-1−/− mice (K. Masuda and H. Kawamoto, unpublished observations). Transfer of Hes-1-null FL cells into Rag2-null host mice failed to generate mature T cells in the thymus, and the expansion of DN cells before TCR gene rearrangement was severely affected (32). In the FTOC system, Hes-1−/− fetal thymocytes were found at a reduced number despite the almost normal development to the DP and SP stages (33). The level of Hes-1 expression is profoundly involved in the cellularity of developing thymocytes, especially for the expansion of early T progenitors. In conclusion, we suggest that down-regulation of the Hes-1 gene is a major cause of the impaired expansion of early T progenitors in mel-18−/− FT.

Because we found a normal proportion of cycling cells and an increased number of TUNEL-positive cells in 12.5 dpc mel-18−/− FT (Fig. 4, A and B), we believe that the reduced cell number in mel-18−/− FT resulted from the increased cell death of early T progenitors. We also found that bcl-2 could partially rescue (50–80%) thymocyte cell number in the absence of mel-18 from the analysis of bcl-2 Tg/mel-18−/− mice (data not shown).

Notch signaling is known to be associated with cell death as well as differentiation and proliferation, although the molecular mechanisms through which Notch activation affects cell death remain to be elucidated (51). In Hes-1−/− mice, the expansion of the earliest thymocytes was severely impaired, but it was not determined whether Hes-1 deficiency affected cell proliferation or increased susceptibility to cell death (32). During normal thymocyte development, immature thymocytes are themselves vulnerable to cell death. It is, therefore, likely that the down-regulation of Hes-1 expression could lead to cell death, even though the cells have the potential to proliferate. However, it is not yet known whether the down-regulation of the Hes-1 gene directly causes thymocyte cell death in mel-18−/− FT or if other mechanisms are involved.

We also investigated whether mechanisms other than the Notch signaling pathway were involved in the dysfunction of early T progenitors in mel-18−/− FT. In a previous report the reduced number of mel-18−/− thymocytes was attributed to their lack of responsiveness to IL-7, based on the observation that total thymocyte of adult mel-18−/− mice displayed a low proliferative response to PMA plus IL-7 (14). In addition, it is well known that IL-7 and SCF play a key role in the proliferation and survival of immature thymocytes (43, 44, 52). Therefore, we thought it possible that an impaired response to IL-7 resulted in the cell death in mel-18−/− fetal T progenitors. However, when we focused on fetal DN1 and DN2 cells, these cells proliferated normally in response to IL-7 or SCF (Fig. 4,C), indicating that the dysfunction of T progenitors in mel-18−/− FT is not associated with IL-7 signal. Several lineage-specific genes (GATA-3, Tcf-1, Lef-1, and E2A), which have been shown to be essential for early thymocyte development (53), were expressed normally in mel-18−/− DN1/DN2 cells (Fig. 5). It was reported that the induction of GATA-3 was affected under Th2-skewed conditions in mel-18−/− naive CD4 T cells (18); however, we observed normal expression of GATA-3 in mutant DN1/DN2 cells. Therefore, GATA-3 was not a target of Mel-18 in the T progenitor population. Furthermore, we suspected that ink4a, another known downstream target gene of the PcG complex, was involved in the dysfunction of mel-18−/− early T progenitors. However, we were not able to detect the expression of the ink4a gene in either mel-18−/− or control DN1/DN2 cells, as examined by RT-PCR Southern hybridization procedure (data not shown). Therefore, it is not likely that the defects in T progenitors in mel-18−/− mice are correlated with the expression of the ink4a gene.

In the TSt-4/DL1 coculture system, we noted that after 6 days, the expression level of Hes-1 in mel-18−/− T precursors was particularly decreased compared with that in mel-18+/− cells, whereas after day 3 the expression of Hes-1 was not affected in mel-18−/− T precursors (Fig. 6,A). In parallel, we noticed that the difference in Hes-1 expression between mel-18+/− and mel-18−/− cells widened as differentiation proceeded from DN1 to DN2 (Fig. 5). From these results, we hypothesize a biphasic regulation of Hes-1 expression. During the early phase, the Hes-1 gene is transactivated by Notch-1 signaling. In the later phase, the Hes-1 expression level is maintained during the expansion of early T progenitors. Regarding the early phase, we examined the possibility of involvement of mel-18 in transactivation of the Hes-1 gene by Notch-1 signals. However, we did not find that mel-18 influenced the Hes-1 promoter activities mediated by Notch-1-IC in a luciferase assay (data not shown). This finding corresponds well with the results from the TSt-4/DL1 assay, in which the Hes-1 expression level in mel-18−/− T precursors after 3 days of culture was comparable to that in control cells. For the later phase, we also observed that the expression of Hes-1 is down-regulated to a remarkably low level when the amount of mel-18 mRNA was reduced by siRNA in the KKC cell line, which expresses the Hes-1 gene in the absence of Notch ligands (Fig. 6 B). These results suggest that mel-18 plays a key role in maintenance of the active state of the Hes-1 gene in proliferating T progenitors.

To maintain gene expression levels in proliferating cell populations, daughter cells must inherit a cellular memory of the state of gene expression from the mother cell after cell division. As shown in Fig. 7, prethymic T progenitors migrate into the thymus, then they interact with stroma cells to receive several signals, including a Notch-1 signal. Prompted by these signals, T progenitors start to proliferate massively and differentiate from DN1 to DN2 cells to create a diversity of TCR β-chains (pre-β proliferation). In mel-18−/− FT, early T progenitors could not proliferate appropriately and were susceptible to cell death, probably due to their inability to maintain the Hes-1 expression level after each cell division.

In Drosophila melanogaster, the expression of HOM-C genes is initially induced by gap and pair-rule genes, and these genes are transiently expressed during early embryogenesis. Subsequently, the Trithorax group (TrxG) and PcG genes are necessary to maintain proper expression of HOM-C throughout development. TrxG proteins are generally responsible for preserving the active state, whereas PcG proteins maintain a silenced state. However, recent observations revealed that some members of the PcG and TrxG groups exhibit the dual property of maintaining both the activated and inactivated states of homeotic gene expression. They are identified as enhancers of Trithorax and Polycomb (ETP) (11), although it remains to be determined whether ETP maintains both states of gene expression directly or indirectly. Interestingly, mel-18 has been classified as a member of the mammalian ETP. In this paper we demonstrate that mel-18, as an ETP, could, in principle, play a positive role in Hes-1 transcriptional maintenance in proliferating early T progenitors (ETPs), i.e., ETP controls expansion of ETPs. However, it is at this time unclear whether mel-18 acts directly on Hes-1 gene expression.

In summary, we demonstrated that the PcG gene mel-18 is indispensable for the expansion of adult and fetal early T progenitors. The PcG/mel-18 protein complex appears to regulate the expression of Hes-1, a target gene of the Notch-1 signaling pathway, in these progenitors. We propose that PcG/mel-18 plays a crucial role in pre-β proliferation, which is required for the formation of a diversified TCR β-chain, through the maintenance of Hes-1 gene expression. Our study indicates that PcG/mel-18, in principle, works as a cellular memory system in pre-β proliferation, thereby supporting the development of the diversified immune system.

The authors have no financial conflict of interest.

We thank Kenji Tanigaki and Tasuku Honjo for providing the vectors used in the luciferase assays, Kiyokazu Kakugawa for providing the TSt-4/DL1 cell line, Akihiko Muto and Tomoo Ueno for technical assistance, and Yoshihiro Takihara, Yosuke Takahama, and Yoshimoto Katsura for helpful discussions and advice.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by Grants-in-Aid for Science from the Ministry of Education, Culture, Sports, Science, and Technology of Japan. The authors have no financial conflict of interest.

3

Abbreviations used in this paper: PcG, Polycomb group; DL1, Delta-like-1; DN, double negative; DP, double positive; dpc, days postcoitum; ETP, enhancers of Trithorax and Polycomb, early T progenitor; FB, fetal blood; FL, fetal liver; FT, fetal thymus; FTOC, FT organ culture; HOM-C, homeotic gene; HOS, high oxygen submersion; HSC, hemopoietic stem cell; KSL, LinScaI+c-Kit+ cell; Lin, lineage marker-negative; PI, propidium iodide; PRC1, Polycomb repressive complex 1; 7R+, LinSca-1+c-KitlowIL-7Rα+; SCF, stem cell factor; siRNA, small interfering RNA; SP, single positive; TrxG, Trithorax group; TSt-4/DL1, TSt-4 stroma cell expressing Delta-like-1.

1
van Lohuizen, M..
1999
. The trithorax-group and Polycomb-group chromatin modifiers: implications for disease.
Curr. Opin. Genet. Dev.
9
:
355
.
2
Orlando, V..
2003
. Polycomb, epigenomes, and control of cell identity.
Cell
112
:
599
.
3
Alkema, M. J., M. Bronk, E. Verhoeven, A. Otte, L. J. van’t Veer, A. Berns, M. van Lohuizen.
1997
. Identification of Bmi1-interacting proteins as constituents of a multimeric mammalian Polycomb complex.
Genes Dev.
11
:
226
.
4
Gunster, M. J., D. P. Satijn, K. M. Hamer, J. L. den Blaauwen, D. de Bruijn, M. J. Alkema, M. van Lohuizen, R. van Driel, A. P. Otte.
1997
. Identification and characterization of interactions between the vertebrate Polycomb-group protein BMI1 and human homologs of polyhomeotic.
Mol. Cell. Biol.
17
:
2326
.
5
Shao, Z., F. Raible, R. Mollaaghababa, J. R. Guyon, C. T. Wu, W. Bender, R. E. Kingston.
1999
. Stabilization of chromatin structure by PRC1, a Polycomb complex.
Cell
98
:
37
.
6
Breiling, A., B. M. Turner, M. E. Bianchi, V. Orlando.
2001
. General transcription factors bind promoters repressed by Polycomb group proteins.
Nature
412
:
651
.
7
Saurin, A. J., Z. Shao, H. Erdjument-Bromage, P. Tempst, R. E. Kingston.
2001
. A Drosophila Polycomb group complex includes Zeste and dTAFII proteins.
Nature
412
:
655
.
8
Cao, R., L. Wang, H. Wang, L. Xia, H. Erdjument-Bromage, P. Tempst, R. S. Jones, Y. Zhang.
2002
. Role of histone H3 lysine 27 methylation in Polycomb-group silencing.
Science
298
:
1039
.
9
Fischle, W., Y. Wang, S. A. Jacobs, Y. Kim, C. D. Allis, S. Khorasanizadeh.
2003
. Molecular basis for the discrimination of repressive methyl-lysine marks in histone H3 by Polycomb and HP1 chromodomains.
Genes Dev.
17
:
1870
.
10
Raaphorst, F. M., A. P. Otte, C. J. Meijer.
2001
. Polycomb-group genes as regulators of mammalian lymphopoiesis.
Trends Immunol.
22
:
682
.
11
Lessard, J., G. Sauvageau.
2003
. Polycomb group genes as epigenetic regulators of normal and leukemic hemopoiesis.
Exp. Hematol.
31
:
567
.
12
van der Lugt, N. M., J. Domen, K. Linders, M. van Roon, E. Robanus-Maandag, H. te Riele, M. van der Valk, J. Deschamps, M. Sofroniew, M. van Lohuizen, et al
1994
. Posterior transformation, neurological abnormalities, and severe hematopoietic defects in mice with a targeted deletion of the bmi-1 proto-oncogene.
Genes Dev.
8
:
757
.
13
Core, N., S. Bel, S. J. Gaunt, M. Aurrand-Lions, J. Pearce, A. Fisher, M. Djabali.
1997
. Altered cellular proliferation and mesoderm patterning in Polycomb-M33-deficient mice.
Development
124
:
721
.
14
Akasaka, T., K. Tsuji, H. Kawahira, M. Kanno, K. Harigaya, L. Hu, Y. Ebihara, T. Nakahata, O. Tetsu, M. Taniguchi, et al
1997
. The role of mel-18, a mammalian Polycomb group gene, during IL-7-dependent proliferation of lymphocyte precursors.
Immunity
7
:
135
.
15
Tokimasa, S., H. Ohta, A. Sawada, Y. Matsuda, J. Y. Kim, S. Nishiguchi, J. Hara, Y. Takihara.
2001
. Lack of the Polycomb-group gene rae28 causes maturation arrest at the early B-cell developmental stage.
Exp. Hematol.
29
:
93
.
16
Jacobs, J. J., B. Scheijen, J. W. Voncken, K. Kieboom, A. Berns, M. van Lohuizen.
1999
. Bmi-1 collaborates with c-Myc in tumorigenesis by inhibiting c-Myc-induced apoptosis via INK4a/ARF.
Genes Dev.
13
:
2678
.
17
Tetsu, O., H. Ishihara, R. Kanno, M. Kamiyasu, H. Inoue, T. Tokuhisa, M. Taniguchi, M. Kanno.
1998
. mel-18 negatively regulates cell cycle progression upon B cell antigen receptor stimulation through a cascade leading to c-myc/cdc25.
Immunity
9
:
439
.
18
Kimura, M., Y. Koseki, M. Yamashita, N. Watanabe, C. Shimizu, T. Katsumoto, T. Kitamura, M. Taniguchi, H. Koseki, T. Nakayama.
2001
. Regulation of Th2 cell differentiation by mel-18, a mammalian Polycomb group gene.
Immunity
15
:
275
.
19
Rodewald, H. R., H. J. Fehling.
1998
. Molecular and cellular events in early thymocyte development.
Adv. Immunol.
69
:
1
.
20
Kawamoto, H., K. Ohmura, S. Fujimoto, M. Lu, T. Ikawa, Y. Katsura.
2003
. Extensive proliferation of T cell lineage-restricted progenitors in the thymus: an essential process for clonal expression of diverse T cell receptor β chains.
Eur. J. Immunol.
33
:
606
.
21
Mallick, C. A., E. C. Dudley, J. L. Viney, M. J. Owen, A. C. Hayday.
1993
. Rearrangement and diversity of T cell receptor β chain genes in thymocytes: a critical role for the β chain in development.
Cell
73
:
513
.
22
Falk, I., J. Biro, H. Kohler, K. Eichmann.
1996
. Proliferation kinetics associated with T cell receptor-β chain selection of fetal murine thymocytes.
J. Exp. Med.
184
:
2327
.
23
Pear, W. S., F. Radtke.
2003
. Notch signaling in lymphopoiesis.
Semin. Immunol.
15
:
69
.
24
Radtke, F., A. Wilson, G. Stark, M. Bauer, J. van Meerwijk, H. R. MacDonald, M. Aguet.
1999
. Deficient T cell fate specification in mice with an induced inactivation of Notch1.
Immunity
10
:
547
.
25
Pui, J. C., D. Allman, L. Xu, S. DeRocco, F. G. Karnell, S. Bakkour, J. Y. Lee, T. Kadesch, R. R. Hardy, J. C. Aster, et al
1999
. Notch1 expression in early lymphopoiesis influences B versus T lineage determination.
Immunity
11
:
299
.
26
Izon, D. J., J. C. Aster, Y. He, A. Weng, F. G. Karnell, V. Patriub, L. Xu, S. Bakkour, C. Rodriguez, D. Allman, et al
2002
. Deltex1 redirects lymphoid progenitors to the B cell lineage by antagonizing Notch1.
Immunity
16
:
231
.
27
Washburn, T., E. Schweighoffer, T. Gridley, D. Chang, B. J. Fowlkes, D. Cado, E. Robey.
1997
. Notch activity influences the αβ versus γδ T cell lineage decision.
Cell
88
:
833
.
28
Wolfer, A., A. Wilson, M. Nemir, H. R. MacDonald, F. Radtke.
2002
. Inactivation of Notch1 impairs VDJβ rearrangement and allows pre-TCR-independent survival of early αβ lineage thymocytes.
Immunity
16
:
869
.
29
Tanigaki, K., M. Tsuji, N. Yamamoto, H. Han, J. Tsukada, H. Inoue, M. Kubo, T. Honjo.
2004
. Regulation of αβ/γδ T cell lineage commitment and peripheral T cell responses by Notch/RBP-J signaling.
Immunity
20
:
611
.
30
Amsen, D., J. M. Blander, G. R. Lee, K. Tanigaki, T. Honjo, R. A. Flavell.
2004
. Instruction of distinct CD4 T helper cell fates by different notch ligands on antigen-presenting cells.
Cell
117
:
515
.
31
Schmitt, T. M., J. C. Zuniga-Pflucker.
2002
. Induction of T cell development from hematopoietic progenitor cells by δ-like-1 in vitro.
Immunity
17
:
749
.
32
Tomita, K., M. Hattori, E. Nakamura, S. Nakanishi, N. Minato, R. Kageyama.
1999
. The bHLH gene Hes1 is essential for expansion of early T cell precursors.
Genes Dev.
13
:
1203
.
33
Kaneta, M., M. Osawa, K. Sudo, H. Nakauchi, A. G. Farr, Y. Takahama.
2000
. A role for pref-1 and HES-1 in thymocyte development.
J. Immunol.
164
:
256
.
34
Kawamoto, H., K. Ohmura, N. Hattori, Y. Katsura.
1997
. Hemopoietic progenitors in the murine fetal liver capable of rapidly generating T cells.
J. Immunol.
158
:
3118
.
35
Egawa, T., K. Kawabata, H. Kawamoto, K. Amada, R. Okamoto, N. Fujii, T. Kishimoto, Y. Katsura, T. Nagasawa.
2001
. The earliest stages of B cell development require a chemokine stromal cell-derived factor/pre-B cell growth-stimulating factor.
Immunity
15
:
323
.
36
Watanabe, Y., O. Mazda, Y. Aiba, K. Iwai, J. Gyotoku, S. Ideyama, J. Miyazaki, Y. Katsura.
1992
. A murine thymic stromal cell line which may support the differentiation of CD48 thymocytes into CD4+8 αβ T cell receptor positive T cells.
Cell. Immunol.
142
:
385
.
37
Allman, D., A. Sambandam, S. Kim, J. P. Miller, A. Pagan, D. Well, A. Meraz, A. Bhandoola.
2003
. Thymopoiesis independent of common lymphoid progenitors.
Nat. Immunol.
4
:
168
.
38
Porritt, H. E., L. L. Rumfelt, S. Tabrizifard, T. M. Schmitt, J. C. Zuniga-Pflucker, H. T. Petrie.
2004
. Heterogeneity among DN1 prothymocytes reveals multiple progenitors with different capacities to generate T cell and non-T cell lineages.
Immunity
20
:
735
.
39
Kondo, M., I. L. Weissman, K. Akashi.
1997
. Identification of clonogenic common lymphoid progenitors in mouse bone marrow.
Cell
91
:
661
.
40
Kawamoto, H., T. Ikawa, K. Ohmura, S. Fujimoto, Y. Katsura.
2000
. T cell progenitors emerge earlier than B cell progenitors in the murine fetal liver.
Immunity
12
:
441
.
41
Ikawa, T., K. Masuda, M. Lu, N. Minato, Y. Katsura, H. Kawamoto.
2004
. Identification of the earliest prethymic T-cell progenitors in murine fetal blood.
Blood
103
:
530
.
42
Itoi, M., H. Kawamoto, Y. Katsura, T. Amagai.
2001
. Two distinct steps of immigration of hematopoietic progenitors into the early thymus anlage.
Int. Immunol.
13
:
1203
.
43
Rodewald, H. R., M. Ogawa, C. Haller, C. Waskow, J. P. DiSanto.
1997
. Pro-thymocyte expansion by c-kit and the common cytokine receptor γ chain is essential for repertoire formation.
Immunity
6
:
265
.
44
von Freeden-Jeffry, U., N. Solvason, M. Howard, R. Murray.
1997
. The earliest T lineage-committed cells depend on IL-7 for Bcl-2 expression and normal cell cycle progression.
Immunity
7
:
147
.
45
Hashimoto, Y., K. J. Blank.
1990
. T cell receptor genes and T cell development in virus-transformed early T cell lines.
J. Immunol.
144
:
1518
.
46
Lessard, J., S. Baban, G. Sauvageau.
1998
. Stage-specific expression of Polycomb group genes in human bone marrow cells.
Blood
91
:
1216
.
47
Ohta, H., A. Sawada, J. Y. Kim, S. Tokimasa, S. Nishiguchi, R. K. Humphries, J. Hara, Y. Takihara.
2002
. Polycomb group gene rae28 is required for sustaining activity of hematopoietic stem cells.
J. Exp. Med.
195
:
759
.
48
Park, I. K., D. Qian, M. Kiel, M. W. Becker, M. Pihalja, I. L. Weissman, S. J. Morrison, M. F. Clarke.
2003
. Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells.
Nature
423
:
302
.
49
Iso, T., L. Kedes, Y. Hamamori.
2003
. HES and HERP families: multiple effectors of the Notch signaling pathway.
J. Cell. Physiol.
194
:
237
.
50
Harman, B. C., E. J. Jenkinson, G. Anderson.
2003
. Entry into the thymic microenvironment triggers Notch activation in the earliest migrant T cell progenitors.
J. Immunol.
170
:
1299
.
51
Miele, L., B. Osborne.
1999
. Arbiter of differentiation and death: Notch signaling meets apoptosis.
J. Cell. Physiol.
181
:
393
.
52
Marsden, V. S., A. Strasser.
2003
. Control of apoptosis in the immune system: Bcl-2, BH3-only proteins and more.
Annu. Rev. Immunol.
21
:
71
.
53
Staal, F. J., F. Weerkamp, A. W. Langerak, R. W. Hendriks, H. C. Clevers.
2001
. Transcriptional control of T lymphocyte differentiation.
Stem Cells
19
:
165
.
54
Hattori, N., H. Kawamoto, S. Fujimoto, K. Kuno, Y. Katsura.
1996
. Involvement of transcription factors TCF-1 and GATA-3 in the initiation of the earliest step of T cell development in the thymus.
J. Exp. Med.
184
:
1137
.