The most immature population of fetal thymus (FT) cells has been shown to generate not only T but also B and myeloid cells. The present study was undertaken to clarify whether such a multipotent activity of the earliest population of FT cells is attributed to multipotent hemopoietic progenitors or to a mixture of lineage-restricted progenitors. Examination of individual FT progenitors by a recently established clonal assay system, which is able to determine the developmental potential of each progenitor toward T, B, and myeloid lineages, elucidated that a large majority of progenitors in FT were restricted to the T cell lineage. Presence of a small number of B or myeloid lineage-restricted progenitors was also disclosed. No multipotent progenitors, however, were detected in FT. These results are consistent with our recent finding that restriction of hemopoietic stem cells to T, B, and myeloid lineages takes place in the fetal liver.
Tcells mainly develop in the thymus, whereas the progenitors of T cells are of extrathymic origin. It has long been disputed whether the prethymic T cell progenitors are the hemopoietic stem cells themselves or are T cell lineage committed (1, 2, 3). If progenitors migrating into the thymus are multipotent, the thymic environment may have to play a dual role that concerns commitment of the multipotent progenitors to the T cell lineage and differentiation of the committed progenitors into T cells. If the T cell lineage commitment occurs prethymically, the major task assigned to the thymus is specialized to support the differentiation and growth of T cells from the committed progenitors.
Using a recently established clonal assay system, named multilineage progenitor (MLP)3 assay (4), which is able to determine the developmental potential of each progenitor toward T, B, and myeloid cells, we have succeeded in obtaining direct evidence for the presence of progenitors (p-T) capable of generating T cells but not B or myeloid cells in the fetal liver (FL). However, the possibility has not been ruled out that multipotent progenitors (p-Multi) or stem cells migrate into the thymus, since the most immature cell population in fetal thymus (FT) is shown to be able to generate not only T but also B and myeloid cells, when cocultured with a deoxyguanosine (dGuo)-treated FT lobe or with stromal cells (5, 6). The present study was performed to clarify the developmental capability of individual cells in the earliest population of FT with the MLP assay system. The results indicated that all progenitors in FT were lineage restricted, of which a large majority were p-T.
Materials and Methods
Pregnant C57BL/6 (B6) mice were purchased from SLC (Shizuoka, Japan), and B6Ly5.1 mice were maintained in our animal facility. FT and FL were obtained from B6Ly5.1 mice at 12 days postcoitum (dpc) and were used as the source of progenitors. B6 fetuses (15 dpc) were used as the source of dGuo-treated FT lobes.
The following Abs were used: anti-Ly5.1 (A20-1.7, donated by Dr. Y. Saga, Banyu Seiyaku, Tokyo, Japan), anti-Ly5.2 (ALI-4A2, donated by Dr. I. L. Weissman, Stanford University, San Francisco, CA), anti-B220 (RA3-6B2, obtained from American Type Culture Collection (ATCC), Manassas, VA), phycoerythrin (PE)-anti-Mac-1 (M1/70, Caltag, South San Francisco, CA), PE-anti-Gr-1 (RA3-8C5, PharMingen, San Diego, CA), PE-anti-B220 (RA3-6B2, Caltag), allophycocyanin (APC)-anti-Thy-1.2 (5a-8, Caltag), anti-CD25 (PC61.5.3, ATCC), FITC-anti-CD45 (30F11.1, PharMingen), PE-anti-CD45 (30F11.1, PharMingen), PE-anti-Sca-1 (E13-161.7, PharMingen), anti-FcγRII/III (FcR) (2.4G2, PharMingen), and anti-c-kit (ACK-2, donated by Dr. S.-I. Nishikawa, Kyoto University, Japan). Anti-Ly5.1 and anti-FcR were labeled with FITC. Anti-c-kit, anti-B220, anti-CD25, and anti-Ly5.2 were labeled with cyanine 5 (Cy5)(Cy5 labeling kit; Biologic Detection Systems, Pittsburgh, PA).
Recombinant murine (rm) IL-7 was kindly donated by Dr. T. Sudo (Basic Research Laboratory, Toray, Kanagawa, Japan). Commercially available rm stem cell factor (SCF) (Genzyme, Cambridge, MA), rm IL-3 (Genzyme), rm granulocyte-macrophage (GM)-CSF (Life Technologies, Gaithersburg, MD), rm IL-1α (Genzyme), recombinant human granulocyte (G)-CSF (Life Technologies), and recombinant human macrophage (M)-CSF (Cellular Products, Buffalo, NY) were also used.
High oxygen submersion (HOS) culture and MLP assay
HOS culture of FT organs was performed as described elsewhere (7, 8). The MLP assay culture, which is able to examine the capability of a progenitor to give rise not only to T but also to B and myeloid cells, is a modification of the HOS culture. The detailed method for MLP assay has previously been described (4). Briefly, single dGuo-treated lobes were submerged in 0.2 ml of RPMI 1640 medium supplemented with FCS (10%), SCF (10 ng/ml), IL-3 (3 ng/ml), and IL-7 (200 U/ml) in a well of a 96-well V-bottom plate (Costar, Cambridge, MA). CD44+CD25−FcR− (FcR−) cells in 12-dpc FT and c-kit+CD45+Sca-1+(Sca-1+) cells in 12-dpc FL were sorted as described (4, 6). Single cells were lifted using a micropipet under direct microscopic visualization and seeded into the wells. Plates were centrifuged at 150 × g for 5 min at room temperature, placed into a plastic bag (Ohmi Oder Air Service, Hikone, Japan), and the air was exchanged with a gas mixture of 5% CO2, 70% O2, and 25% N2. The plastic bag was then incubated at 37°C.
Cells grown both inside and outside the cultured lobe were harvested from each well after 10 days of culture, and single cell suspensions were made. One third of each cell sample was stained with FITC-anti-Ly5.1 and Cy5-anti-Ly5.2, and analyzed by flow cytometry. The samples containing Ly5.1+ cells were selected for further analysis. The remaining two thirds of cells from the selected samples were divided into two batches. One batch was stained with FITC-anti-Ly5.1, PE-anti-B220, and allophycocyanin-anti-Thy1.2, and the other stained with FITC-anti-Ly5.1, PE-anti-Mac-1, PE-anti-Gr-1, and Cy5-anti-B220. The flow cytometric analysis of surface phenotypes is detailed in the previous paper (4).
Long-term culture initiation assay
FT or FL cells were plated on confluent layers of stromal cell line FBMD-1 (9) in 48-well plates coated with rat tail collagen type I (Becton Dickinson Labware, Bedford, MA). The culture medium was IMDM (Life Technologies) containing 10% FCS, 5% horse serum, and hydrocortisone (10−5 M). The plates were incubated at 33°C in 10% CO2. On day 35, the number of colonies with or without cobblestone area (CA) (10) were counted. To assess long-term culture-initiating cells (LTC-IC), cells in wells that contained colonies were trypsinized and plated individually in semisolid medium to examine the presence of colony forming unit-culture (CFU-C) (11). Wells with replatable CFU-C were scored as LTC-IC. In CFU-C assay, α-MEM (Life Technologies) containing 30% FCS, 1% methylcellulose, 1% BSA, 2-ME (5 × 10−5 M), l-glutamine (1 mM), SCF (10 ng/ml), IL-3 (10 ng/ml), granulocyte-macrophage-CSF (10 ng/ml), IL-1α (10 ng/ml), granulocyte-CSF (10 ng/ml), and macrophage-CSF (10 U/ml) were used. After 14 days of incubation, colonies consisting of more than 50 cells were scored as CFU-C.
Results and Discussion
Investigation of the developmental capability of thymic progenitors with the MLP assay
All lymphoid cells in 12-dpc FT are Thy-1−CD4−CD8−CD3−c-kit+CD45+CD44+CD25− (CD44+CD25−), which is known to be the most immature subset of thymocytes (12). Our previous investigation indicated that FcR− subpopulation of the CD44+CD25− FT cells (Fig. 1) retains the multipotent progenitor activity (6). A total of 150 individual FcR− FT cells from 12-dpc fetuses of the B6Ly5.1 strain were cultured under MLP assay conditions. Cells were harvested on the 10th day from each well and assayed for expression of T, B, and myeloid cell markers. The type of progenitors seeded in each well was determined from the phenotype of the cells grown in the well, and representative flow cytometric profiles are shown in Figure 2,B. p-T, and progenitors restricted to B (p-B) and myeloid (p-M) lineages, were detected, but no p-Multi or bipotent progenitors were found. Profiles of cells derived from p-T, p-B, p-M, and p-Multi of FL are also shown (Fig. 2 C). Thymic p-T are able to produce a large number of T cells, almost to the same extent as FL p-T. On the other hand, thymic p-B and p-M generated much smaller numbers of cells than p-B and p-M in FL.
In Figure 3 (left panel), numbers of different types of progenitors detected among 150 FcR− FT cells, as well as the estimated total numbers of these progenitors per a pair of FT lobes, are shown. Growth of T, B, and myeloid cells was seen in 31, 2, and 3 wells, respectively, whereas no wells were seen in which two or three lineage cells were generated. Because a pair of FT lobes from a 12-dpc fetus contains about 400 FcR− cells, the total numbers of p-T, p-B, and p-M were calculated as approximately 83, 5, and 8, respectively. These values are very similar to the numbers of progenitors estimated by limiting dilution analyses, in which it has been found that progenitors generating T, B, and myeloid cells in 12-dpc FT are 73, 6, and 15, respectively (our unpublished data). These results strongly suggest that the previous finding, that the earliest population of FT cells is able to generate T, B, and myeloid cells (5, 6), is attributable to the fact that it contains a mixture of different lineage-restricted progenitors but not to the presence of p-Multi.
For comparison, progenitors in the c-kit+Sca-1+ population of FL (see Figure 1) were also examined. p-Multi as well as lineage-restricted progenitors were detected in the FL (Fig. 3, right panel). Irrespective of the lineage, the number of progenitors is much higher in FL than in FT. For example, 20 times higher number of p-T are present in FL. We have also found that a large number of p-T of the same phenotype as those in FL are present in the blood of 12-dpc fetuses (about 600 per fetus).4 These p-T, however, are phenotypically distinct from the T cell progenitors previously reported to be present in the blood of 15.5-dpc fetuses, which are Thy-1+c-kitlow (13).
Although progenitors generating T and B cells, or p-TB, have been reported to be present in adult bone marrow (14), we have not succeeded in detecting such p-TB type progenitors in either FT or FL. As we have discussed elsewhere (4), it is probable that the frequency of p-TB in FL, should they exist, is too low to be detected, or that the p-TB stage is skipped over in the hemopoiesis in fetuses.
Further investigations for searching earlier progenitors or stem cells in FT
The following experiments were conducted to verify with different systems the finding made with the MLP assay (Fig. 3) that no p-Multi are present in FT. Early hemopoietic progenitors are known to form long lasting colonies with CA (10) on a stromal cell monolayer, and the most primitive progenitors or stem cells can be detected as LTC-IC (11). A large number of nonfractionated FT cells (12 dpc), equivalent to 3 FT lobes, were assayed for CA-forming cells (CAFC) and LTC-IC activities. As shown in Table I, only three CAFC were detected among 1,440 FT cells, and no LTC-IC were found. This is in marked contrast with the detection of 37 CAFC and 24 LTC-IC among 144 Sca-1 high positive (Sca-1high) FL cells (prepared as in Figure 1), where p-Multi are highly enriched.4
|Cell Typea .||No. of Cells Assayed .||No. of Colonies Detected .||.||.|
|.||.||Non-CAFCb .||CAFCb .||LTC-IC\E .|
|Cell Typea .||No. of Cells Assayed .||No. of Colonies Detected .||.||.|
|.||.||Non-CAFCb .||CAFCb .||LTC-IC\E .|
Ten FT cells or one FL Sca-1high cell from 12-dpc fetuses of B6Ly5.1 mice were seeded with a micropipet onto the monolayer of stromal cell line FBMD-1 in three 48-well plates. Total number of FT cells and Sca-1high FL cells per a 12-dpc fetus are approximately 1000 and 5000, respectively.
The numbers of colonies without or with cobblestone area (non-CAFC or CAFC, respectively) were scored on day 35.
We next examined whether FT cells were able to reconstitute hemopoiesis of irradiated recipients. Lethally irradiated B6 mice rescued with 105 B6 bone marrow cells were transferred i.v. with 3,000 12-dpc FT cells from B6Ly5.1 fetuses. Control mice were given 100 Sca-1high FL cells. After 4 mo, no donor type cells were detected in the blood of 7 of 7 recipients of FT cells, whereas all 4 recipients of FL Sca-1high cells contained donor type cells (5–17%) in their peripheral blood. The results are consistent with the findings that no p-Multi (Fig. 3) or LTC-IC (Table I) exist in FT.
Characterization of the progenitors migrating into the thymus to produce T cells has been one of the most basic problems in the study of the mechanisms of T cell development. The present results, in combination with our previous findings (4) that restriction of stem cells to T, B, and myeloid lineages occurs in FL (see also Fig. 3 of this paper), strongly suggested that it is the p-T that migrates into the thymus.
It has been shown that NK cells and lymphoid dendritic cells are derived from thymic progenitors (15, 16, 17). Since the MLP assay system used in the present investigation is not contrived for determining the differentiation and/or growth of these cells, it is unclear whether the thymic p-T, p-B, and p-M determined here are exactly unipotent or are also able to generate NK or dendritic cells. So far, we have not yet tried to modify the MLP assay cultures to generate dendritic cells. On the other hand, very recently it became possible to induce NK cells in thymic organ cultures (HOS culture) by adding IL-2, and we found that a large majority of p-T in FcR− FT cells retained the capability of giving rise to NK cells (our unpublished observation). It is not yet known, however, whether other types of progenitors such as p-M, p-B, or prethymic p-T are also able to generate NK cells. More detailed investigation is necessary to clarify the exact process of commitment, including NK and DC lineages.
We are indebted to Dr. W. T. V. Germeraad (University Hospital Utrecht, The Netherlands) for reading the manuscript, and to Ms. Y. Takaoki for secretarial assistance.
This study was partially supported by grants from the Ministry of Education, Science, Sports and Culture, Japan, and from the Japan Health Sciences Foundation.
Abbreviations used in this paper: MLP, multilineage progenitor; CA, cobblestone area; CAFC, cobblestone area-forming cells; CFU-C, CFU-culture; dGuo, deoxyguanosine; dpc, days postcoitum; FcR, FcγRII/III; FL, fetal liver; FT, fetal thymus; HOS culture, high oxygen submersion culture; LTC-IC, long-term culture-initiating cells; p-B, progenitors capable of generating B cells but not T or myeloid cells; p-M, progenitors capable of generating myeloid cells but not T or B cells; p-Multi, multipotent progenitor; p-T, progenitors capable of generating T cells but not B or myeloid cells; B6, C57BL/6; PE, phycoerythrin; Cy5, cyanine 5; rm, recombinant murine; SCF, stem cell factor.
H. Kawamoto, K. Ohmura, S. Fujimoto, and Y. Katsura. Submitted for publication.