We investigated the developmental potential of hemopoietic progenitors in the aorta-gonad-mesonephros (AGM) region, where the definitive type hemopoietic progenitors have been shown to emerge before the fetal liver develops. By using an assay system that is able to determine the developmental potential of individual progenitors toward the T, B, and myeloid lineages, we show that not only multipotent progenitors but also progenitors committed to the T, B, or myeloid lineage already exist in this region of day 10 fetuses. Bipotent progenitors generating myeloid and T cells or those generating myeloid and B cells were also detected, suggesting that the commitment to T and B cell lineages is in progress in the AGM region. The numbers of these progenitors, however, were only 1/200–1/1000 of those in fetal liver of day 12 fetuses. Such small numbers of progenitors suggest that hemopoiesis has just started in the AGM region of day 10 fetuses. Although most of T cell lineage-committed progenitors in the AGM region generated only a small number of immature T cells, some were able to generate a large number of mature T cells. The detection of various types of lineage-committed progenitors strongly suggests that the AGM region is not only the site of stem cell emergence, but also the site of hemopoiesis, including lineage commitment. The T cell progenitors found in the AGM region may represent the first immigrants to the thymus anlage.

It has been shown that hemopoietic progenitors are present in the embryo body (EB)3 before fetal liver (FL) hemopoiesis begins (1, 2). In the EB, hemopoietic progenitors are first observed in the paraaortic splanchnopleura (P-Sp) (2, 3), and the site of production of the progenitors shifts from P-Sp to the aorta-gonad-mesonephros (AGM) region by 10 days postcoitum (dpc) (4). The presence of hemopoietic stem cells (HSC) with long-term repopulating (LTR) potential has been demonstrated in the AGM region of 10.5-dpc fetuses (5). On the other hand, the progenitors in the yolk sac (YS), which had been proposed to be the tissue where HSC originate (6), are considered to represent a primitive type participating in the hemopoiesis of early embryos (7, 8, 9), although recent studies indicated that YS progenitors were able to acquire LTR potential when they were injected directly into the liver of neonatal mice (10, 11).

It is still controversial when in the fetus the progenitors capable of generating lymphocytes emerge, despite extensive investigations having been undertaken on the differentiational potential of early progenitors in the EB. Ogawa and colleagues (12) have previously shown that B cell progenitors emerge in the 9.6-dpc EB by culturing whole embryo cells on a monolayer of a bone marrow (BM)-derived stromal cell line. On the other hand, Cumano et al. (3) showed that progenitors in 8.5-dpc (10–12 somite stage) P-Sp generated B cells when they were cultured on a stromal cell monolayer in the presence of cytokines, including SCF and IL-7, and moreover that cells derived from a single 9.5-dpc P-Sp cell precultured on the stromal cell layer for 14 days showed T cell progenitor activity when transferred into a deoxyguanosine (dGuo)-treated fetal thymus (FT) lobe (13). Recently, we found that the progenitors in the caudal half of 9.5-dpc EB were able to generate both T and B cells (14), whereas 8.5-dpc (5–10 somite stage) embryos did not contain T or B cell progenitors. Thus, a dramatic change of the characteristics of progenitors, such as commitment to T or B cell lineage, has occurred at about 9.5 dpc.

The presence of B cell lineage-committed progenitors (p-B) has been elucidated in the FL at 12 dpc by the group of Cumano (15, 16). In their experimental system, however, T cell lineage-committed progenitors (p-T) cannot be detected, because p-T do not differentiate on a stromal cell monolayer. Nor can it be clarified whether the progenitors generating only B cells, only myeloid cells, or both B and myeloid cells on the stromal cell layer retained T cell-generating ability. To broaden the area of analysis, we have recently established a clonal assay system, named multilineage progenitor (MLP) assay, which is effective in determining the developmental potential of individual progenitors toward T, B, and myeloid cell lineages (17). By using this assay system, we have shown the presence of various types of progenitors, including multipotent progenitors (p-Multi), p-T, p-B, and myeloid lineage-committed progenitors (p-M) in 12-dpc FL (17, 18). We have also shown that it is p-T, but not p-Multi, that migrate to the thymus to produce T cells (18, 19). However, since the thymic lymphoid cells appear at 11 dpc (20), it is probable that the first p-T migrating into the FT may have emerged somewhere other than FL before 11 dpc. In the present study, we show that lineage-committed progenitors as well as p-Multi are present in the AGM region of 10-dpc fetuses.

Adult C57BL/6 (B6) mice were purchased from Japan SLC (Shizuoka, Japan). B6Ly 5.1 mice were maintained in our animal facility. (B6 × B6Ly 5.1)F1 fetuses were obtained by crossing the male B6Ly 5.1 mice with female B6 mice. Embryos at various stages of gestation were obtained from time-mated pregnant mice. The date of finding the vaginal plug was designated as 0 dpc. Precise embryonic stages were determined by counting the pairs of somites. Embryos with 24–29 pairs of somites were regarded as 9.5 dpc, and those with 30–35 pairs of somites were regarded as 10.0 dpc of gestation.

AGM regions, fetal blood (FB), FL, and FT were obtained from B6Ly 5.1 fetal mice or (B6 × B6Ly 5.1)F1 fetuses. Single cell suspensions of the AGM cells were prepared by passage of tissues containing AGM region through a 26-gauge needle. FB was collected by cutting the umbilical cord without damaging the embryo. Embryos were left bleeding in the medium. To get rid of maternal blood, the uterus and embryos had been extensively washed.

The following Abs were used: FITC anti-Mac-1 (M1/70; Caltag, San Francisco, CA), FITC anti-Gr-1 (RA3-8C5; PharMingen, San Diego, CA), FITC anti-B220 (RA3-6B2; Caltag), FITC anti-Thy-1.2 (5a-8; Caltag), FITC anti-CD8 (YTS169.4; Caltag), FITC anti-mouse IgM (Cappel, West Chester, PA), FITC anti-TCRγδ (GL-3; Caltag), FITC anti-CD3ε (145-2C11; PharMingen), FITC anti-Vβ6 (RR4-7; PharMingen), FITC anti-Vβ8 (KJ16; Caltag), FITC anti-Vγ3 (536; PharMingen), PE anti-B220 (RA3-6B2; Caltag), PE anti-CD45 (30F11.1; PharMingen), PE anti-Thy-1.2 (5a-8; Caltag), PE anti-CD4 (GK1.5; Caltag), PE anti-TCRαβ (H57-597; Caltag), PE anti-Sca-1 (E13-161.7; PharMingen). Anti-TER119 (TER) (established by Dr. T. Kina in our laboratory) and anti-Ly-5.1 (A20-1.7, donated by Dr. Y. Saga, Banyu Seiyaku, Tokyo, Japan) were labeled with FITC in our laboratory. Anti-c-kit (ACK-2, donated by Dr. S.-I. Nishikawa, Kyoto University, Kyoto, Japan) and anti-Ly-5.1 were conjugated with cyanine 5 (Cy5) using a labeling kit (Biological Detection Systems, Pittsburgh, PA). Anti-TER, anti-Mac-1, anti-Gr-1, anti-B220, and anti-Thy-1.2 were used as lineage markers (Lin).

The basic procedures for HOS culture have been described previously (21). Briefly, FT obtained from 15-dpc fetuses of B6 mice were treated with dGuo. The lobes were washed, and single dGuo-treated lobes were placed into wells of a V-bottom 96-well plate (Costar, Cambridge, MA), into which a suspension of progenitor cells was added. The plates were sealed in a plastic bag (Ohmi Oder Air Service, Hikone, Japan) and the air inside was replaced by a gas mixture (70% O2, 25% N2, and 5% CO2). The plastic bag was incubated at 37°C. RPMI 1640 medium supplemented with 10% FCS, l-glutamine (2 mM), sodium pyruvate (1 mM), sodium bicarbonate (2 mg/ml), nonessential amino acid solution (0.1 mM), 2-ME (5 × 10−5 M), streptomycin (100 μg/ml), and penicillin (100 U/ml) was used as complete medium. Medium change was performed every 4 to 5 days.

The MLP assay culture has been designed as a modification of the HOS culture (17). Cells sorted with a flow cytometer were suspended in a 9-cm plastic dish (INA · OPTICA, Osaka, Japan) at the concentration of about 100 cells/ml. A single cell was picked up using a micropipette with a long fine tip (QSP, Petalurna, CA) under direct microscopic visualization, and was put into each well of a V-bottom 96-well plate in which a dGuo-treated FT lobe had been placed in complete medium. A mixture of cytokines including 10 ng/ml murine rSCF (Genzyme, Cambridge, MA), 200 U/ml murine IL-7 (donated by Dr. T. Sudo, Basic Research Laboratory, Toray, Kanagawa, Japan), and 1 ng/ml murine GM-CSF (Life Technologies, Gaithersburg, MD) was added to the culture to promote the growth of B and myeloid cells. The difference of the cytokine mixture used here from that used for the MLP assay for FL progenitors (17) is that GM-CSF was used in place of IL-3. This change is due to the fact that most progenitors in the AGM region tend to become blastic and do not differentiate in the presence of IL-3 (our unpublished data). Plates were centrifuged, placed in a plastic bag, and cultured under HOS conditions at 37°C. Half of the medium was replaced with complete medium containing 200 U/ml IL-7 and 1 ng/ml GM-CSF on the fifth day.

The stromal cell line TSt-4 was used to investigate the development of B and myeloid cells, as described previously (22). To a confluent monolayer of TSt-4 cells in a six-well plate (Costar), FL or AGM cells were added. The culture medium is the same complete medium as used in HOS cultures, except that the concentration of FCS is 5%.

The basic methods for cell surface staining and analysis have been described previously (23). Cells obtained from the AGM region, FB, FL, FT, and BM were stained with various mAbs and were sorted using a FACS Vantage. The sorted cells were reanalyzed to check their purity, and were found to be >98% pure.

In the MLP assay, cells generated inside and outside the FT lobe were harvested from each well 10 days after cultivation, and single cell suspensions were made. Half of the cell sample was stained with FITC anti-B220, PE anti-Thy-1.2, and Cy5 anti-Ly-5.1, whereas the other half was stained with FITC anti-Mac-1, FITC anti-Gr-1, PE anti-B220, and Cy5 anti- Ly 5.1. In some experiments, flow-cytometric analysis was performed by using half of each sample, while the remaining half was used for PCR analysis. Surface phenotypes were analyzed with a FACS Vantage by using CELL Quest software (version 1.2.2). Cells showing the phenotype of Ly 5.1+Thy-1+B220 or Ly-5.1+Thy-1B220+ were tentatively regarded as donor-derived T or B cells, respectively. These populations were then checked for forward-side scatter and expression of Mac-1/Gr-1, and the cells falling within the lymphocyte area and expressing neither Mac-1 nor Gr-1 were finally considered to be either donor-derived T or B cells. Ly 5.1+Thy-1B220Mac-1+Gr-1+ cells were judged as donor-derived myeloid cells. Although this population contains cells with low to middle levels of staining for B220 or Thy-1, such a staining was regarded as nonspecific because these Mac-1+/Gr-1+ cells were found to fall in the area of macrophages in forward-side scatter analysis.

RNA isolation.

mRNA was isolated from TERc-kit+Ly 5.1+ cells (3000 cells) in 10-dpc AGM region and FB of (B6 × B6Ly 5.1)F1 fetuses, Linc-kit+CD45+ cells in 12-dpc FL and FT cells (3000 cells each) of B6 fetuses, and c-kit+CD45+Sca-1high cells in 12-dpc FL cells (3000 cells) of B6 fetuses by using a QuickPrep Micro mRNA Purification Kit (Pharmacia, Little Chalfont, U.K.).

Reverse transcription.

A mixture of mRNA solution and 5 μl of 0.04 μg/μl oligo(dT) primers (Life Technologies) was incubated at 65°C for 5 min. Samples were placed on ice and the following reagents were added: 4 μl of 5× reverse-transcription buffer (0.25 M Tris-HCl, pH 8.3, 0.375 M KCl, and 15 mM MgCl2), 2 μl of 0.1 M DTT, 0.4 μl of 25 mM dNTPs, and 100 U M-MLV reverse transcriptase (Life Technologies). The reaction samples were incubated at 37°C for 60 min and inactivated at 70°C for 5 min, then chilled on ice.

PCR.

cDNA was amplified by PCR using various sets of primers. Primers used: TCF-1 sense, 5′-CCAGCTTTCTCCACTCTACG-3′; TCF-1 antisense, 5′-TCAAGGATGGGTGGGTGAAC-3′; GATA-3 sense, 5′-TCGGCCATTCGTACATGGAA-3′; GATA-3 antisense, 5′-GAGAGCCGTGGTGGATGGAC-3′; mb-1 sense, 5′-ATCATCTTGCTGTTCTGTGC-3′; mb-1 antisense, 5′-ACACTAACGAGGATGCTGTA-3′; c-fms sense, 5′-CTGGAGAAGAAATATGTGCG-3′; c-fms antisense, 5′-TTCAGACCAAGCGAGAAGAT-3′; IL-7R sense, 5′-AAGGATGTGGTGAATGCAGG-3′; IL-7R antisense, 5′-ACAATAGGACAGGTTCATGG-3′; β-actin sense, 5′-TCCTGTGGCATCCATGAAACT-3′; β-actin antisense, 5′-GAAGCACTTGCGGTGCACGAT-3′. The reaction volume was 20 μl containing 2 μl of cDNA sample, 2 μl of 10× PCR buffer (0.1 M Tris-HCl, pH 9, 0.5 M KCl, and 15 mM MgCl2), 0.16 μl of 25 mM dNTPs, 0.6 U Taq polymerase (Pharmacia), and 0.4 μl of each primer (10 mM). After incubation for 5 min at 94°C, PCR amplification was performed using the Thermal-Cycler (TAKARA, Otsu, Japan). Cycling times and temperatures were as follows: denaturation at 94°C for 1 min, annealing at 55°C for 1 min, and elongation at 72°C for 2 min. Amplification was performed for 20 cycles for β-actin and 30 cycles for all other genes. Fifteen microliters of PCR product were electrophoresed through 1.2% agarose or 5% polyacrylamide gel and stained with ethidium bromide.

After 10 days of MLP assay culture, half of each sample was used for flow-cytometric analysis, and the remaining half was subjected to PCR analysis. Cells (3000 cells) were resuspended in 20 μl of 1× PCR buffer (10 mM Tris-HCl, pH 9, 50 mM KCl, and 1.5 mM MgCl2) including 0.45% Nonidet P-40, 0.45% Tween 20, and 1.2 μg proteinase K (Sigma, St. Louis, MD), and incubated at 55°C for 1 h, then 95°C for 10 min. Samples of these disrupted cells were used as templates for PCR amplification. Primers were: Dβ2, 5′-GCACCTGTGGGGAAGAAACT-3′; Jβ2.6, 5′-TGAGAGCTGTCTCCTACTATCGATT-3′; Vγ4, 5′-AGTGTTCAGAAGCCCGATGCA-3′; Jγ1, 5′-AGAGGGAATTACTATGAGCT-3′. The reaction volume was 20 μl containing 5 μl of the cell extract (equivalent to 750 cells), 1.5 μl of 10× PCR buffer, 0.16 μl of 25 mM dNTPs, 0.4 μl of each primer (10 mM), and 0.6 U Taq polymerase. Thermocycling conditions were as follows: 5 min at 94°C, followed by 35 cycles of 1 min at 94°C, 1 min at 60°C for D-Jβ rearrangement or at 55°C for V-Jγ rearrangement, 2 min at 72°C, and 10 min at 72°C. Amplified DNA products were applied to a 1.2% agarose gel, electrophoresed, and stained with ethidium bromide.

We first investigated when progenitors capable of generating T cells emerge in the AGM region. Unfractionated cells from the AGM region of 9.5-dpc fetuses (4–10 × 104 cells; equivalent to 2–5 fetuses) or those of 10.0-dpc fetuses (3 × 104 cells; equivalent to 1 fetus) were cultured together with a dGuo-treated FT lobe, and the cells generated in the wells were assayed with a flow cytometer. It was found that T cell generation was always observed in cultures of 10.0-dpc AGM cells, whereas in cultures of 9.5-dpc AGM cells, T cell generation was seen only in two of seven experiments (data not shown). These results suggested that progenitors capable of generating T cells emerged between 9.5 and 10.0 dpc.

We investigated the time course of T cell generation from 10.0-dpc AGM progenitors. Unfractionated AGM cells (3 × 104 cells per well), and as controls 12-dpc FL cells (3 × 103 cells per well) and Linc-kit+ adult BM cells (2 × 103 cells per well), were cultured together with a dGuo-treated FT lobe. The number of cells used in each group is set in order that the numbers of T cell progenitors included in these groups are comparable. Cells grown in each well were harvested at different days, enumerated, and analyzed by flow-cytometric analysis. Numbers of Thy-1+ cells per well are plotted in Fig. 1,A, and CD4 vs CD8 and TCRαβ vs TCRγδ expression profiles of cells generated in these cultures are shown in Fig. 1,B. T cell generation from AGM progenitors was much more rapid than that from BM progenitors (Fig. 1,A), although it was slower by about 2 days than that from 12-dpc FL progenitors. The time course of T cell generation from AGM progenitors seems to conform to the profile seen in ontogenic T cell development in FT (18, 24). It was also found that the proportions of Vβ6+, Vβ8+, or Vγ3+ cells among T cells generated from the 10.0-dpc AGM progenitors were comparable to those among T cells generated from 12-dpc FL progenitors (Table I). From these results, the T cell progenitors in the AGM region may have been qualified as those migrating into FT.

FIGURE 1.

T cell generation from the AGM progenitors. Unfractionated cells from the AGM region of 10.0-dpc fetuses (3 × 104), and for comparison unfractionated cells from FL of 12-dpc fetuses (3 × 103) and Linc-kit+ adult BM cells (2 × 103) were cultured with a dGuo-treated FT lobe in wells of a 96-well V-bottom plate under HOS conditions. At various days of culture, cells were collected from each lobe, counted, and analyzed with a flow cytometer. The results are representative of three independent experiments. A, Numbers (geometric mean and SD of five wells) of Thy-1+ cells per lobe are plotted. B, CD4 vs CD8 and TCRαβ vs TCRγδ profiles of representative samples are shown.

FIGURE 1.

T cell generation from the AGM progenitors. Unfractionated cells from the AGM region of 10.0-dpc fetuses (3 × 104), and for comparison unfractionated cells from FL of 12-dpc fetuses (3 × 103) and Linc-kit+ adult BM cells (2 × 103) were cultured with a dGuo-treated FT lobe in wells of a 96-well V-bottom plate under HOS conditions. At various days of culture, cells were collected from each lobe, counted, and analyzed with a flow cytometer. The results are representative of three independent experiments. A, Numbers (geometric mean and SD of five wells) of Thy-1+ cells per lobe are plotted. B, CD4 vs CD8 and TCRαβ vs TCRγδ profiles of representative samples are shown.

Close modal
Table I.

Expression of Vβ6, Vβ8, and Vγ3 on T cells developed from 10.0-dpc AGM and 12-dpc FLa

Vβ6Vβ8Vγ3
AGM 12.3b 18.2 13.5 
FL 14.6 18.8 15.7 
Vβ6Vβ8Vγ3
AGM 12.3b 18.2 13.5 
FL 14.6 18.8 15.7 
a

Unfractionated 10.0-dpc AGM cells (3 × 104) and 12-dpc FL cells (3 × 103) were cultured together with a dGuo-treated FT lobe under HOS conditions. On the 16th day of culture, cells were harvested and stained with anti-TCR mAb for flow cytometric analysis. The data are representative of two independent experiments.

b

Percentage in CD3+ cells.

We next examined the ability of AGM progenitors to generate B and myeloid cells. When unfractionated cells from 9.5-dpc AGM region (1 × 105 cells) were cultured on a monolayer of the stromal cell line TSt-4, which supports the development of both B and myeloid cells, myeloid cell generation was constantly observed, whereas B cell generation was only occasionally seen (data not shown) like the T cell generation in FT organ cultures (see preceding section). On the other hand, 10.0-dpc AGM cells (3 × 104 unfractionated cells) constantly produced B and myeloid cells. In the following experiments, we used 10.0-dpc fetuses as the source of AGM progenitors.

AGM cells (3 × 104 unfractionated cells) from 10.0-dpc fetuses, and for comparison 12-dpc FL cells (3 × 103 unfractionated cells), were cultured on a monolayer of TSt-4, and floating cells were harvested on days 5, 10, and 14. The recovered cells were assayed for expression of B220, Mac-1/Gr-1, and surface IgM (Fig. 2). At an early phase (day 5), the culture of AGM progenitors was predominated by myeloid cells. This is in contrast with the fact that the cultures of FL progenitors were dominated by B cells. B cell growth from AGM progenitors became prominent on day 10 of culture (data not shown), and on day 14 the number of B cells became comparable with that in cultures of FL progenitors. The B cells derived from 10.0-dpc AGM progenitors express surface IgM within 14 days after initiation of the culture. It was also found that these B cells were negative for surface CD5 (data not shown). The results in this and the preceding section indicate that T cell progenitors emerge nearly at the same gestational age as B cell progenitors.

FIGURE 2.

Generation of B and myeloid cells from AGM progenitors. Unfractionated cells from the AGM region of 10.0-dpc fetuses (3 × 104), and those from FL of 12-dpc fetuses (3 × 103) were cultured on a monolayer of the stromal cell line TSt-4 in a six-well plate. Floating cells were collected 5 and 14 days later, stained with various mAb as indicated in the figure, and analyzed with a flow cytometer. The results are representative of three independent experiments.

FIGURE 2.

Generation of B and myeloid cells from AGM progenitors. Unfractionated cells from the AGM region of 10.0-dpc fetuses (3 × 104), and those from FL of 12-dpc fetuses (3 × 103) were cultured on a monolayer of the stromal cell line TSt-4 in a six-well plate. Floating cells were collected 5 and 14 days later, stained with various mAb as indicated in the figure, and analyzed with a flow cytometer. The results are representative of three independent experiments.

Close modal

AGM and FB cells from 10.0-dpc fetuses, and also FL and FT cells from 12-dpc fetuses, were stained with various mAb and analyzed by flow cytometry. Flow-cytometric profiles of AGM, FB, and FL cells are shown in Fig. 3,A. Previous investigations indicated that progenitors are enriched in the Sca-1+ population in FL and FB of 12–13-dpc fetuses (18, 25), but no Sca-1-expressing cells are found in the AGM region or FB of 10.0-dpc fetuses (Fig. 3 A). On the other hand, it seemed probable that hemopoietic progenitors are enriched in the TERc-kit+CD45+ population, because the TERc-kit+CD45 population contains angioblasts (26) and probably also primordial germ cell progenitors (27).

FIGURE 3.

Flow-cytometric profiles of cells in the AGM region, FB, and FL (A), and expression of hemopoietic lineage-associated genes in these cells (B). A, Cells from the AGM region and FB (10.0-dpc) as well as FL (12 dpc) were stained with various mAbs, and analyzed with a FACS Vantage. The numbers of cells in TERc-kit+CD45+ population in the AGM region and FB of a 10.0-dpc fetus are 600 and 2,000, respectively, and the number of Linc-kit+CD45+ cells in 12-dpc FL is 45,000. B, mRNA was prepared from cells gated with rectangles in A and Linc-kit+CD45+ FT cells (12 dpc). cDNA was prepared by reverse transcription, and the PCR products were electrophoresed and stained with ethidium bromide. GATA-3 (28 ) and TCF-1 (29 ) are known to be associated with T cell lineages, whereas IL-7R is essential for T and B cell development (30 ). mb-1 (31 ) and c-fms (32 ) are associated with B and myeloid cell lineage, respectively. AGM c-kit and FB c-kit represent the TERc-kit+CD45+ cells of 10-dpc AGM region and FB, respectively. FL c-kit and FT c-kit represent the Linc-kit+CD45+ cells of 12-dpc FL and FT, respectively. FL Sca-1hi represents Linc-kit+Sca-1high cells of 12-dpc FL. Two independent experiments showed a similar expression pattern.

FIGURE 3.

Flow-cytometric profiles of cells in the AGM region, FB, and FL (A), and expression of hemopoietic lineage-associated genes in these cells (B). A, Cells from the AGM region and FB (10.0-dpc) as well as FL (12 dpc) were stained with various mAbs, and analyzed with a FACS Vantage. The numbers of cells in TERc-kit+CD45+ population in the AGM region and FB of a 10.0-dpc fetus are 600 and 2,000, respectively, and the number of Linc-kit+CD45+ cells in 12-dpc FL is 45,000. B, mRNA was prepared from cells gated with rectangles in A and Linc-kit+CD45+ FT cells (12 dpc). cDNA was prepared by reverse transcription, and the PCR products were electrophoresed and stained with ethidium bromide. GATA-3 (28 ) and TCF-1 (29 ) are known to be associated with T cell lineages, whereas IL-7R is essential for T and B cell development (30 ). mb-1 (31 ) and c-fms (32 ) are associated with B and myeloid cell lineage, respectively. AGM c-kit and FB c-kit represent the TERc-kit+CD45+ cells of 10-dpc AGM region and FB, respectively. FL c-kit and FT c-kit represent the Linc-kit+CD45+ cells of 12-dpc FL and FT, respectively. FL Sca-1hi represents Linc-kit+Sca-1high cells of 12-dpc FL. Two independent experiments showed a similar expression pattern.

Close modal

Expression of several genes that are closely associated with T, B, or myeloid lineage was investigated in the TERc-kit+CD45+ cells from 10.0-dpc AGM region and FB. As controls, cell populations from FL and FT of 12-dpc fetuses exhibiting the same phenotypes (Linc-kit+CD45+) as well as the Sca-1 high positive (Sca-1high) population of Linc-kit+ cells from 12-dpc FL, in which p-Multi are enriched (18), were used. In this experiment, (B6 × B6Ly 5.1)F1 fetuses from a pregnant B6 mother (Ly 5.2) were used, because it is possible to obtain cell populations of fetuses devoid of cells derived from maternal blood simply by gating out Ly 5.1 cells. mRNA was extracted from 3000 cells. These mRNA samples were reverse transcribed to cDNA, and one-tenth served for PCR amplification of each gene. Expression levels of GATA-3, mb-1, c-fms, and IL-7R in the c-kit+CD45+ cells of AGM and FB populations were comparable to those in the corresponding population of FL (Fig. 3 B). In contrast, the bands representing the products of mb-1, c-fms, and IL-7R in the Sca-1high cells of 12-dpc FL were faint or almost invisible. These findings strongly suggested that in the AGM region, various types of committed progenitors are present in the c-kit+CD45+ population. The failure to detect any TCF-1 transcripts except for FT may indicate that T cell differentiation has not yet begun in the AGM region nor in the FB.

In the present study, we succeeded in applying the MLP assay system to investigate the developmental potential of individual progenitors in the AGM region. This was attained by changing the combination of cytokines added to the HOS culture (see Materials and Methods). Single cells of the TERc-kit+CD45+ AGM population obtained from 10.0-dpc fetuses of B6Ly-5.1 or (B6 × B6Ly 5.1)F1 mice were cultured together with a dGuo-treated lobe in the presence of SCF, GM-CSF, and IL-7 (Fig. 4 A). On the tenth day of culture, cells were harvested from each well, stained with various mAb, and flow cytometrically analyzed.

FIGURE 4.

Representative flow-cytometric profiles of cells generated from a p-T1, p-T2, p-Multi1, and p-Multi2 in the MLP assay. A, Single cell manipulation and MLP assay culture. Cells sorted with a flow cytometer are placed in a 9-cm dish. Single cells are picked up under microscopic visualization using a pipette with a long fine tip, and transferred into each well of a V-bottom 96-well plate in which a dGuo-treated FT lobe had been placed. A mixture of cytokines was added to the culture medium, and the cells were cultured under high oxygen submersion conditions at 37°C. B, Comparison of cells generated from a p-T1 and a p-T2 with those from a p-T of 12-dpc FL. C, Comparison of cells generated from a p-Multi1 and p-Multi2 with those from a p-Multi of 12-dpc FL. Cells were harvested on the tenth day of culture. As progenitor source, (B6 × B6Ly 5.1)F1 and B6Ly 5.1 fetuses were used in experiments shown in B and C, respectively. In the culture of FL p-Multi, generation of DP cells is usually observed on day 14 of culture (data not shown).

FIGURE 4.

Representative flow-cytometric profiles of cells generated from a p-T1, p-T2, p-Multi1, and p-Multi2 in the MLP assay. A, Single cell manipulation and MLP assay culture. Cells sorted with a flow cytometer are placed in a 9-cm dish. Single cells are picked up under microscopic visualization using a pipette with a long fine tip, and transferred into each well of a V-bottom 96-well plate in which a dGuo-treated FT lobe had been placed. A mixture of cytokines was added to the culture medium, and the cells were cultured under high oxygen submersion conditions at 37°C. B, Comparison of cells generated from a p-T1 and a p-T2 with those from a p-T of 12-dpc FL. C, Comparison of cells generated from a p-Multi1 and p-Multi2 with those from a p-Multi of 12-dpc FL. Cells were harvested on the tenth day of culture. As progenitor source, (B6 × B6Ly 5.1)F1 and B6Ly 5.1 fetuses were used in experiments shown in B and C, respectively. In the culture of FL p-Multi, generation of DP cells is usually observed on day 14 of culture (data not shown).

Close modal

Six different types of progenitors were observed, which were p-Multi, p-T, p-B, p-M, and bipotent progenitors generating myeloid and T cells (p-MT) and those generating myeloid and B cells (p-MB). Common lymphoid progenitors (p-TB) were not detected. Because these six types are principally the same as those found in FL (17, 18), we do not show the flow-cytometric profiles of cells generated from all six types. On the other hand, we depicted profiles exhibiting differences from those of FL progenitors in Fig. 4. It was found that p-T in the AGM region comprise two types, p-T1 and p-T2 (Fig. 4,B). p-T2 are similar to the p-T in FL, because this type of p-T can generate a large number (>20,000 cells) of Thy-1+ cells, and a portion of these cells expresses CD4 and/or CD8 (Fig. 4 B, extreme right panels) as well as TCR (not shown) on their surface. The other type (p-T1), which comprises the majority of p-T in the AGM region, generates only 500-3000 Thy-1+ cells, and these Thy-1+ cells do not express other T cell markers, nor the NK cell marker NK 1.1 (not shown), even after an extended culture period. Removal of GM-CSF from the culture medium did not influence the phenotype of Thy-1+ cells derived from AGM p-T1 (data not shown).

It was also found that two types of p-Multi (p-Multi1 and p-Multi2) exist in the AGM region (Fig. 4,C). p-Multi1-type progenitors generate a small number (<5000 cells) of Thy-1+CD4CD8 cells by day 10 of culture, whereas p-Multi2-type progenitors are capable of generating a large number of T cells expressing CD4 and/or CD8 within 10 days of culture. p-Multi2-type progenitors have never been detected in 12-dpc FL (17, 18 , and our unpublished data) or in the 11-dpc AGM region (not shown). Because a large majority of p-T in the AGM region are p-T1 type, it is likely that the rather rapid generation of mature T cells from unfractionated AGM cells seen in Fig. 1 is at least partially contributed by the p-Multi2.

We next investigated whether the Thy-1+ cells generated from these p-T1 are authentic T cells. DNA was extracted from half of Thy-1+ cells derived from two p-T1 and one p-T2, and served for PCR analysis to determine the Dβ2-Jβ2 and Vγ3/Vγ4-Jγ1 rearrangements. T cells derived from a p-Multi1 of the AGM region were also investigated. As positive and negative controls for PCR, 16-dpc FT cells and cells derived from a p-M, respectively, were used. Results are shown in Fig. 5. FT cells from 16-dpc fetuses show six D-Jβ-rearranged bands and a germline band (1.8 kb) as well as V-Jγ-rearranged bands (Vγ4Jγ1 and Vγ3Jγ1). These six D-Jβ- and two V-Jγ-rearranged bands were observed in all samples examined in this study, including the case of 2 p-T1, indicating that cells generated from the p-T1 are surely T cells, and that both p-T1 and p-T2 extensively proliferated before rearrangement of TCR genes.

FIGURE 5.

The rearrangement of TCRβ-chain and TCRγ-chain genes in cells derived from single AGM p-T and p-Multi. Genomic DNA was prepared from cells derived from p-T1, p-T2, and p-Multi1 and as controls, cells derived from an AGM p-M as well as 16-dpc FT cells. The sample (equivalent to 750 cells each) was PCR amplified with the primers shown on the right of the figure, electrophoresed through a 1.2% agarose gel, and stained with ethidium bromide.

FIGURE 5.

The rearrangement of TCRβ-chain and TCRγ-chain genes in cells derived from single AGM p-T and p-Multi. Genomic DNA was prepared from cells derived from p-T1, p-T2, and p-Multi1 and as controls, cells derived from an AGM p-M as well as 16-dpc FT cells. The sample (equivalent to 750 cells each) was PCR amplified with the primers shown on the right of the figure, electrophoresed through a 1.2% agarose gel, and stained with ethidium bromide.

Close modal

Fig. 6 scores the numbers of different types of progenitors found in 582 TERc-kit+CD45+ AGM cells, and 200 TERc-kit+CD45+ FB cells from 10.0-dpc fetuses. Progenitors detected among 234 Linc-kit+CD45+ FL cells from 12-dpc fetuses are also shown. Numbers of progenitors per AGM region, FB, or FL of one fetus can be read on the bottom scale. The proportions of six different types of progenitors in the AGM region are not so much different from those in FL. However, the numbers of progenitors in the AGM region were much smaller than those in FL. For example, the number of p-Multi in the AGM region was only 8, whereas about 2000 p-Multi are present in FL. Such a large difference, however, simply reflects the difference in cell numbers between the 10.0-dpc AGM region and 12-dpc FL. Although not indicated in Fig. 6, among 4 p-T in the AGM region, 3 were p-T1 type and 1 was p-T2 type, and among 8 p-Multi in the AGM region, 6 were p-Multi1 type and 2 were p-Multi2 type.

FIGURE 6.

The frequency and total number of different types of progenitors in the AGM region (10 dpc), FB (10 dpc), and FL (12 dpc). The numbers of different types of progenitor detected among 582 TERc-kit+CD45+ AGM cells, 200 TERc-kit+CD45+ FB cells, and 234 Linc-kit+CD45+ FL cells are scored. In these panels, the sum of p-Multi1 and p-Multi2, and the sum of p-T1 and p-T2 are scored as p-Multi and p-T, respectively. These results are the accumulated data from 10 independent experiments. Calculation of the total numbers of progenitors in the AGM region, FB, and FL is based on the cell number shown in the legend to Fig. 3.

FIGURE 6.

The frequency and total number of different types of progenitors in the AGM region (10 dpc), FB (10 dpc), and FL (12 dpc). The numbers of different types of progenitor detected among 582 TERc-kit+CD45+ AGM cells, 200 TERc-kit+CD45+ FB cells, and 234 Linc-kit+CD45+ FL cells are scored. In these panels, the sum of p-Multi1 and p-Multi2, and the sum of p-T1 and p-T2 are scored as p-Multi and p-T, respectively. These results are the accumulated data from 10 independent experiments. Calculation of the total numbers of progenitors in the AGM region, FB, and FL is based on the cell number shown in the legend to Fig. 3.

Close modal

As in FB of 12-dpc fetuses (18), a large majority of progenitors in 10.0-dpc FB are lineage committed. Of special interest is that the number of p-M in FB is much larger than that in the AGM region. This is in marked contrast with the finding that the number of p-M in 12-dpc FB is much smaller than that in 12-dpc FL (18). One p-MT and p-T were found in 200 c-kit+ cells of 10.0-dpc FB, and this FB p-T was p-T2 type. Other types of progenitors such as p-B or p-Multi were not detected. The failure to detect any p-B or p-Multi, however, does not necessarily indicate that these progenitors do not circulate, but that the number of circulating p-B or p-Multi is very small (less than 10 progenitors per FB).

The presence of various types of progenitors, p-T, p-B, p-M, and p-Multi, as well as bipotent progenitors p-MT and p-MB, which are thought to be on the process of lineage commitment, was disclosed in the 10.0-dpc AGM region. These six types of progenitors are the same as those observed in 12-dpc FL, although some of the progenitors in the AGM region seemed to be of a “primitive” type.

Characteristics of the progenitors in AGM region are slightly different from those in FL. For example, AGM progenitors do not express Sca-1, which is the marker of the earliest progenitors in FL (18, 25). GM-CSF, but not IL-3, was effective in supporting the differentiation of AGM progenitors toward myeloid lineage, although FL progenitors respond to both IL-3 and GM-CSF (17). With IL-3, AGM progenitors tend to become blastoid without generating T, B, or myeloid cells. These differences may reflect the difference in developmental stage of progenitors obtained at different embryonic ages as well as the in vivo environment they belonged to. Of interest is the fact that the same six types of progenitors as found in FL (17, 18) were identified in the AGM region, regardless of such differences in surface phenotype and cytokine responsiveness. This may indicate that hemopoietic stem cells or multipotent progenitors begin lineage commitment in the AGM region before completing the maturational steps of the multipotent progenitor itself. Conforming to the finding in FL (17, 18), all bipotent progenitors detected in the AGM region were p-MT and p-MB types, and p-TB-type progenitors were not detected. These results strongly suggested that commitment to p-T and p-B in the AGM region occurs mainly through bipotent p-MT and p-MB progenitors, respectively. Although p-TB-type progenitors have been reported to be present in the Linc-kitlowIL-7R+ population of adult BM (33), a recent report from the same research group documented that the p-TB-type progenitors may not exist in FL (34). Our present study strongly suggested that the process of hemopoietic cell differentiation in the AGM region is similar to that in FL, but not BM.

We have previously reported the existence of p-T in 12-dpc FL (17, 18). The present work is the first to identify p-T in 10-dpc fetuses, at the gestational age when the thymus has not yet developed. We consider that the progenitors detected as p-T are T cell lineage restricted in vivo, because of the following reasons. First, the MLP assay is effective in discriminating p-T from p-Multi in FL, as detailed in our previous study (18, 19). Second, it was found that the AGM p-T are self reproductive or proliferate before the rearrangement of TCRβ gene without producing other lineage cells (Fig. 5). Such a self-reproducing capability may also indicate that the lineage commitment is a very early event in hemopoiesis. Third, RT-PCR analysis indicated that the transcription factor GATA-3 that is shown to be specific to T cell lineage (28) was expressed in the c-kit+ AGM cells (Fig. 3,B). The expression of IL-7R that is essential to early differentiation of T and B lineage cells is also seen in the c-kit+ AGM cells (Fig. 3 B) as in c-kit+ FL cells, in which a large number of p-T were present (18).

It was also found that unipotent progenitors are heterogeneous in their differentiational potential. For example, p-T can be divided into p-T1 and p-T2 (Fig. 4,B). Among a total of 19 p-T to date detected in the AGM region, 17 were p-T1 type and 2 were p-T2 type. In contrast, only 1 p-T detected in the 10.0-dpc FB was p-T2 type. In 11-dpc AGM region, 2 of 7 p-T were p-T2 type (data not shown), indicating that the proportion of p-T2 increases with the fetal age. Such p-T1-type progenitors, however, have not ever been observed in 12-dpc FL (data not shown). It has not yet been clarified, however, whether p-T1 are simply the immature type of p-T that could mature into p-T2 if maintained in the environment of the AGM region, or whether they are the progenitors of a different type of T cells, such as those developing extrathymically. On the other hand, it is improbable that p-T1 is restricted to the NK lineage, because TCR gene rearrangement is seen in the progeny of p-T1 (Fig. 5).

It was surprising that p-Multi2-type progenitors were discovered in the AGM region, which generate T cells as rapidly as p-T2 (Fig. 4). It is highly probable that p-T2-type progenitors are derived from such p-Multi2. More intensive investigation, however, is required to clarify the maturational steps of hemopoietic stem cells themselves as well as the mechanism of lineage commitment.

The number of progenitors capable of generating B cells, which is the total number of p-Multi, p-MB, and p-B in the AGM region of a 10.0-dpc fetus (30–35 somite stage) detected by our MLP assay, was 10 (Fig. 6). This value is equivalent to the number of the B cell progenitors (15 cells) in a 9.5-dpc P-Sp (25 somite stage) previously estimated by culturing single cells on the stromal cell monolayer (13), indicating that the seeding efficiency in the MLP assay cultures is comparable to that in cultures on stromal cells. On the other hand, the MLP assay is much more efficient in detecting progenitors than the CFU-S assay. It has been shown that day 8 CFU-S and day 11 CFU-S observed in the 10-dpc AGM region were only 0.8 and 1.2, respectively (1, 35), and a large proportion of these CFU-S was erythroid progenitors (35). With the MLP assay, the total number of progenitors capable of generating myeloid cells (the sum of p-Multi, p-MT, p-MB, and p-M) was 38.

The numbers of progenitors in the 10-dpc AGM region are much smaller than those in 12-dpc FL, the numbers of p-Multi, p-T, p-B and p-M in the AGM region being 1/250, 1/1000, 1/400, and 1/200 of those in 12-dpc FL, respectively. If all FL progenitors are derived from the AGM region, extensive proliferation and/or self-reproduction of p-Multi, as well as delivery of committed progenitors should have occurred either or both in the AGM region and FL. Immigration from the YS may also contribute to the definitive hemopoiesis in FL (10, 11).

Generation of T and B cells from 9.5-dpc AGM cells was observed only in two of seven experiments, despite the fact that a large number of cells, equivalent to two to five fetuses, were seeded in each culture. On the other hand, 10.0-dpc AGM progenitors always generated T and B cells. Garcia-Porrero et al. (36) showed by microscopic observation that the formation of the hemopoietic clusters on the vascular endothelial cell layer of the AGM region starts at 9.5 dpc. Our recent work further suggested that the main lymphoid progenitors in 9.5-dpc fetuses are the vascular endothelial cadherin-positive cells (14). These results strongly suggest that before 9.5 dpc, the progenitors generating lymphocytes are the hemangioblasts that can give rise to both hemopoietic cells and vascular endothelial cells, and that multipotent progenitors restricted to hemopoietic lineages emerge in the AGM region at about 9.5 dpc. Unstable generation of T and B cells from 9.5-dpc AGM progenitors may be attributable to incomplete commitment of the hemangioblast to the hemopoietic lineage. On the other hand, conflicting results about the generation of T and B cells from the progenitors in 8.5–9.5-dpc fetuses in other papers (12, 13, 14, 37, 38) could be due to a difference in cell preparation. Vigorous pipetting or enzymatic cell separation may have released a portion of the hemangioblasts from the endothelial cell layer.

Another important finding is that a large majority of hemopoietic progenitors in 10.0-dpc FB are p-M, and the number of p-M in 10.0-dpc FB reached nearly 1/5 of that in 12-dpc FB (18), despite the fact that the total number of all types of progenitors in FB was 1/250 of that in 12-dpc FB (Fig. 6). Our preliminary experiments showed that at least 1/5 of the p-M in the AGM region can also generate erythrocytes. Circulation of a relatively large number of p-M in 10.0-dpc fetuses may reflect the preferential requirement of phagocytes and erythrocytes during the earlier developmental stages of the embryo.

Because the p-Multi are capable of generating T, B, and myeloid cells, it is probable that at least a portion of the p-Multi represents the hemopoietic stem cells. Although we have not yet completed establishing an MLP assay that covers the erythroid lineage, preliminary experiments indicated that only a small portion of p-Multi in the AGM region generates erythrocytes, thus suggesting that a large majority of p-Multi are restricted to myeloid, T, and B cell lineages (p-MTB). The presence of such p-MTB-type progenitors has also been reported recently by Lacaud et al. (39). On the other hand, we do not have any evidence to regard the AGM p-Multi that are able to generate erythrocytes as the conventional stem cells, because LTR-HSC activity has not been detected in 10-dpc AGM cells (5, 27). If the p-Multi in the AGM region do not have LTR-HSC activity, the AGM p-Multi could represent a primitive type multipotent progenitor giving rise to the stem cells. Studies are in progress to purify the p-Multi in 10.0-dpc AGM region and examine the LTR-HSC activity.

We are indebted to Dr. W. T. V. Germeraad (University Hospital, Utrecht, The Netherlands) for correcting English, and to Ms. Y. Takaoki for secretarial assistance.

1

This study was partially supported by grants from the Ministry of Education, Science, Sports, and Culture, and the Research Grant for Longevity Sciences from the Ministry of Health and Welfare, Japan.

3

Abbreviations used in this paper: EB, embryo body; AGM, aorta-gonad-mesonephros; BM, bone marrow; dGuo, deoxyguanosine; dpc, days postcoitum; FB, fetal blood; FL, fetal liver; FT, fetal thymus; HOS, high oxygen submersion; HSC, hemopoietic stem cell; Lin, lineage markers; LTR, long-term repopulating; MLP, multilineage progenitor; p-B, progenitors committed to B cell lineage; p-M, progenitors committed to myeloid lineage; p-MB, bipotent progenitors capable of generating myeloid and B cells; p-MT, bipotent progenitors capable of generating myeloid and T cells; p-Multi, multipotent progenitors; P-Sp, paraaortic splanchnopleura; p-T, progenitors committed to T cell lineage; SCF, stem cell factor; YS, yolk sac.

1
Medvinsky, A. L., N. L. Samoylina, A. M. Müller, E. A. Dzierzak.
1993
. An early pre-liver intraembryonic source of CFU-S in the developing mouse.
Nature
364
:
64
2
Godin, I. E., P. J. Garcia, A. Coutinho, L. F. Dieterlen, M. A. Marcos.
1993
. Para-aortic splanchnopleura from early mouse embryos contains B1a cell progenitors.
Nature
364
:
67
3
Cumano, A., C. Furlonger, C. J. Paige.
1993
. Differentiation and characterization of B-cell precursors detected in the yolk sac and embryo body of embryos beginning at the 10- to 12-somite stage.
Proc. Natl. Acad. Sci. USA
90
:
6429
4
Robb, L..
1997
. Hematopoiesis: origin pinned down at last?.
Curr. Biol.
7
:
R10
5
Müller, A. M., A. Medvinsky, J. Strouboulis, F. Grosveld, E. Dzierzak.
1994
. Development of hematopoietic stem cell activity in the mouse embryo.
Immunity
1
:
291
6
Moore, M. A., D. Metcalf.
1970
. Ontogeny of the haemopoietic system: yolk sac origin of in vivo and in vitro colony forming cells in the developing mouse embryo.
Br. J. Haematol.
18
:
279
7
Dzierzak, E., A. Medvinsky.
1995
. Mouse embryonic hematopoiesis.
Trends Genet.
11
:
359
8
Cumano, A., F. Dieterlen-Lievre, I. Godin.
1996
. Lymphoid potential, probed before circulation in mouse, is restricted to caudal intraembryonic splanchnopleura.
Cell
86
:
907
9
Medvinsky, A., E. Dzierzak.
1996
. Definitive hematopoiesis is autonomously initiated by the AGM region.
Cell
86
:
897
10
Yoder, M. C., K. Hiatt.
1997
. Engraftment of embryonic hematopoietic cells in conditioned newborn recipients.
Blood
89
:
2176
11
Yoder, M. C., K. Hiatt, P. Dutt, P. Mukherjee, D. M. Bodine, D. Orlic.
1997
. Characterization of definitive lymphohematopoietic stem cells in the day 9 murine yolk sac.
Immunity
7
:
335
12
Ogawa, M., S. Nishikawa, K. Ikuta, F. Yamamura, M. Naito, K. Takahashi, S.-I. Nishikawa.
1988
. B cell ontogeny in murine embryo studied by a culture system with the monolayer of a stromal cell clone, ST2: B cell progenitor develops first in the embryonal body rather than in the yolk sac.
EMBO J.
7
:
1337
13
Godin, I., L. F. Dieterlen, A. Cumano.
1995
. Emergence of multipotent hematopoietic cells in the yolk sac and paraaortic splanchnopleura in mouse embryos, beginning at 8.5 days postcoitus.
Proc. Natl. Acad. Sci. USA
92
:
773
14
Nishikawa, S.-I., S. Nishikawa, H. Kawamoto, H. Yoshida, M. Kizumoto, H. Kataoka, Y. Katsura.
1998
. In vitro generation of lymphohematopoietic cells from endothelial cells purified from murine embryos.
Immunity
8
:
761
15
Cumano, A., C. J. Paige, N. N. Iscove, G. Brady.
1992
. Bipotential precursors of B cells and macrophages in murine fetal liver.
Nature
356
:
612
16
Delassus, S., A. Cumano.
1996
. Circulation of hematopoietic progenitors in the mouse embryo.
Immunity
4
:
97
17
Kawamoto, H., K. Ohmura, Y. Katsura.
1997
. Direct evidence for the commitment of hematopoietic stem cells to T, B and myeloid lineages in murine fetal liver.
Int. Immunol.
9
:
1011
18
Kawamoto, H., K. Ohmura, S. Fujimoto, Y. Katsura.
1999
. Emergence of T cell progenitors without B cell or myeloid differentiation potential at the earliest stage of hematopoiesis in the murine fetal liver.
J. Immunol.
162
:
2725
19
Kawamoto, H., K. Ohmura, Y. Katsura.
1998
. Presence of progenitors restricted to T, B, or myeloid lineage, but absence of multipotent stem cells, in the murine fetal thymus.
J. Immunol.
161
:
3799
20
Owen, J. J. T., M. A. Ritter.
1969
. Tissue interaction in the development of thymus lymphocytes.
J. Exp. Med.
129
:
431
21
Watanabe, Y., Y. Katsura.
1993
. Development of T cell receptor αβ-bearing T cells in the submersion organ culture of murine fetal thymus at high oxygen concentrations.
Eur. J. Immunol.
23
:
200
22
Hattori, N., H. Kawamoto, Y. Katsura.
1996
. Isolation of the most immature population of murine fetal thymocytes that includes progenitors capable of generating T, B and myeloid cells.
J. Exp. Med.
184
:
1901
23
Katsura, Y., T. Kina, Y. Takaoki, S.-I. Nishikawa.
1988
. Quantification of the progenitors for thymic T cells in various organs.
Eur. J. Immunol.
18
:
889
24
Pardoll, D. M., B. J. Fowlkes, J. A. Bluestone, A. Kruisbeek, W. L. Maloy, J. E. Coligan, R. H. Schwartz.
1987
. Differential expression of two distinct T-cell receptors during thymocyte development.
Nature
326
:
79
25
Morrison, S. J., H. D. Hemmati, A. M. Wandycz, I. L. Weissman.
1995
. The purification and characterization of fetal liver hematopoietic stem cells.
Proc. Natl. Acad. Sci. USA
92
:
10302
26
Nishikawa, S.-I., S. Nishikawa, M. Hirashima, N. Matsuyoshi, H. Kodama.
1997
. Progressive lineage analysis by cell sorting and culture identifies FLK1+ VE-cadherin+ cells at a diverging point of endothelial and hematopoietic lineages.
Development
125
:
1747
27
Sánchez, M. J., A. Holmes, C. Miles, E. Dzierzak.
1996
. Characterization of the first definitive hematopoietic stem cells in the AGM and liver of the mouse embryo.
Immunity
5
:
513
28
Ting, C.-N., M. C. Olson, K. P. Barton, J. M. Leiden.
1996
. Transcription factor for development of the T-cell lineage.
Nature
384
:
474
29
Oosterwegel, M., M. van de Wetering, D. Dooijes, L. Klomp, A. Winoto, K. Georgopoulos, F. Meijlink, H. Clevers.
1991
. Cloning of murine TCF-1, a T cell-specific transcription factor interacting with functional motifs in the CD3-ε and T cell receptor enhancers.
J. Exp. Med.
173
:
1133
30
Peschon, J. J., P. J. Morrissey, K. H. Grabstein, F. J. Ramsdell, E. Maraskovsky, B. C. Gliniak, L. S. Park, S. F. Ziegler, D. E. Williams, C. B. Ware, et al
1994
. Early lymphocyte expansion is severely impaired in interleukin 7 receptor-deficient mice.
J. Exp. Med.
180
:
1955
31
Li, Y. S., K. Hayakawa, R. R. Hardy.
1993
. The regulated expression of B lineage associated genes during B cell differentiation in bone marrow and fetal liver.
J. Exp. Med.
178
:
951
32
Sherr, C. J..
1990
. Colony-stimulating factor-1 receptor.
Blood
75
:
1
33
Kondo, M., I. L. Weissman, K. Akashi.
1997
. Identification of clonogenic common lymphoid progenitors in mouse bone marrow.
Cell
91
:
661
34
Akashi, K., D. Traver, M. Kondo, I. L. Weissman.
1999
. Lymphoid development from hematopoietic stem cells.
Int. J. Hematol.
69
:
217
35
Medvinsky, A. L., O. I. Gan, M. L. Semenova, N. L. Samoylina.
1996
. Development of day-8 colony-forming unit-spleen hematopoietic progenitors during early murine embryogenesis: spatial and temporal mapping.
Blood
87
:
557
36
Garcia-Porrero, J. A., I. E. Godin, F. Dieterlen-Lièvre.
1995
. Potential intraembryonic hemogenic sites at pre-liver stages in the mouse.
Anat. Embryol.
192
:
425
37
Liu, C. P., R. Auerbach.
1991
. In vitro development of murine T cells from prethymic and preliver embryonic yolk sac hematopoietic stem cells.
Development
113
:
1315
38
Palacios, R., B. A. Imhof.
1993
. At day 8–8.5 of mouse development the yolk sac, not the embryo proper, has lymphoid precursor potential in vivo and in vitro.
Proc. Natl. Acad. Sci. USA
90
:
6581
39
Lacaud, G., L. Carlsson, G. Keller.
1999
. Identification of a fetal hematopoietic precursor with B cell, T cell, and macrophage potential.
Immunity
9
:
827