We have used a c-kit-specific monoclonal antibody, immuno-fluorescence staining and flow fluorocytometry or microscopy analysis to assess the cell surface expression of the c-kit receptor on a panel of non-trans-formed clones representing different stages of T- and B-lymphocyte development, freshly isolated lymphoid cells from thymus, bone marrow and spleen of young adult C57BL/6 mice and cells from yolk sac, thymus and liver of developing C57BL/6 mouse embryos. Pro-T, Pro-B and Pre-B clones derived from thymus or liver of 14-day embryos are c-kit+. Starting at day 8 to 8.5 in yolk sac, day-10 in fetal liver, and day 11 to 12 in fetal thymus, there are many c-kit+ cells. The number of c-kit+ cells in liver and thymus increases up to day 15 and progressively decreases thereafter. Cell sorter purified c-kit+ day 14 fetal liver cells fully reconstitute the T and B cell compartments of immunodeficient Scid mice. Stromal cells or epithelial cells derived from fetal thymus or liver, which can support growth and differentiation of c-kit+ lymphocyte progenitor clones, synthesize mRNA for Steel Factor (SF), the ligand of c-kit. In the adult mouse, however, c-kit expression is restricted to very early stages of T- and B-lymphocyte development (multipotent progenitors, B-cell/myelocytic progenitors, Pro-T and Pro-B lymphocyte progenitors).

Most cells at the Pre-T, Pre-B and later stages of development do not bear detectable c-kit. Using Cos-1 cells tranfected with mouse SF-cDNA and an antagonistic c-kit receptor-specific antibody, we show that the c-kit/SF system contributes to the survival of lymphocyte progenitors and enhances the proliferative responses of these cells to other growth factors (i.e. IL2, IL3, IL4, IL7). However, the c-kit receptor/SF ligand pair is neither sufficient nor necessary for the differentiation of lymphocyte progenitors into mature T-or B-lymphocytes. Finally, in stromal cell lines from fetal liver and adult bone marrow and thymic epithelial cell lines the level of steady state SF-RNA transcripts is inversely correlated with that of IL-7-mRNA. Moreover, IL7 inhibits the synthesis of SF-mRNA in stromal cells and rIL6 abrogates this inhibitory effect of rIL7. Thus, the expression of SF in stromal cells is subjected to complex regulation by other cytokines produced by the same stromal cells or by neighboring cells in a given microenvironment. The results strongly suggest that the c-kit/SF system plays an important role in the very early stages of development of lymphocytes in the mouse.

Mice carrying mutations at the Steel and dominant white spotting (W) loci have abnormal development of primordial germ cells, neural-crest-derived melanoblasts, haemopoietic stem cells, and erythroid and mast-cell lineages (reviewed in Russel, 1979; Witte 1990). The W gene encodes a cell surface tyrosine kinase receptor, called c-kit (Chabot et al., 1988; Geissler et al., 1988) and the Steel gene encodes the ligand for c-kit, a polypeptide called here Steel Factor (SF) and also referred to as mast-cell growth factor or stem cell factor by others (Zsebo et al., 1990; Huang et al., 1990; Williams et al., 1990). SF can be expressed both in membrane-bound and soluble forms. In vivo the membrane-bound form is probably the physiologically relevant ligand for c-kit (Flanagan et al., 1991).

The expression and function of the c-kit receptor/SF ligand pair in the development of primordial germ cells, melanoblasts, erythrocytes and mast cells have been the subject of intensive study by several groups. In all these cell lineages, evidence has been obtained that this receptor/growth factor pair supports cell survival and facilitates the response to other growth factors (Matsui et al., 1990; Nishikawa et al., 1991; Ogawa et al., 1991; Migliaccio et al., 1991; Orr-Urtreger et al., 1990; McNiece et al., 1991a; Broxmeyer et al., 1991; Keshet et al., 1991; Dolci et al., 1991). Various cell types in the developing nervous system express RNA transcripts for the Steel and the W genes

(Keshet et al., 1991; Orr-Urtreger et al., 1990; Matsui et al., 1990), but the biological significance of this is unclear. Both Steel and W mouse mutants do not show gross abnor-malities in the nervous system (Russel, 1979).

Very little is known about the expression and potential functions of the c-kit/SF system in the development of lymphocytes in the embryo and adult (McNiece et al., 1991b; Palacios and Samaridis, 1992). The experiments reported here aimed to determine the cell surface expression of c-kit protein on cells of the lymphocyte compartments both in the developing embryo and in young adult normal mice and to assess the effects of SF on growth and/or differentiation of T- and B-lymphocyte progenitors. We have used the c-kit-specific monoclonal antibody ACK2 (Nishikawa et al., 1991), immuno-fluorescence staining and flow fluo-rocytometry (FACS) or microscopy analysis to determine the cell surface expression of the c-kit protein on (1) a panel of non-transformed cell lines representing different stages of T- and B-lymphocyte development, (2) freshly isolated lymphoid cells from the thymus, bone marrow and spleen of young adult C57BL/6 mice, and (3) cells from yolk sac, liver and thymus of developing C57BL/6 mouse embryos. We then directly tested the contribution of c-kit/SF to growth and/or differentiation of lymphocyte progenitor clones by using Cos-1 cells stably expressing a mouse cDNA encoding the membrane form of the Steel gene product and a c-kit-receptor-specific antibody able to block the function of c-kit/SF system. Finally, we also studied the synthesis of SF-mRNA in stromal cell lines from fetal liver and bone marrow as well as in thymic epithelial cell lines, with known capacity to support growth/differentiation of B-or T-lymphocyte progenitors in vitro. The results of these experiments are the subject of the present communication.

Mice

C57BL/6, CBA/Thy1a and C.B.17 Scid mice are bred and main-tained in the animal barrier facility of the Basel Institute for Immunology, Basel. C.B.17 Scid mice (8- to 12-week-old female and male) without detectable serum Ig were used. The age of the embryos was determined by scoring the day of the appearance of vaginal plugs, which is taken as day 0. Further assessment of the age of gestation was based on the following criteria; 8 days, a well-developed ectoplacental cone and early membrane formation; 9 days, a well developed yolk sac, a conspicuous heart and no liver pigment; 10 days, early pigmentation in the liver; 11 days, a well-pigmented liver and the absence of a differentiated thymus; 13- to 17-days, the relative size and development of the fetus. As there were variations in the individual development of the embryos within a single pregnancy, the estimates in some instances may be in error by 1/2 to 1 day.

Cell lines

The development and characterization of the bone marrow stromal cell lines RP.0.10 and RP.0.16, the fetal liver-derived stromal line FLS4.1 and the thymic epithelial cell lines ET and EA2, were described (Palacios et al., 1989a,b; Gutierrez and Palacios, 1991; Palacios and Samaridis, 1992). The growth factor-dependent clones: multipotent progenitors (PR-5, PR-8, PR-23) (Palacios, R. and Samaridis, J., submitted for publication), the B-cell/myelocytic progenitors (LyD9, LyB9, LyH7), the marrow Pro-B lymphocyte progenitors (Bc/Bm11, CB/Bm7), the fetal liver Pro-B lymphocyte progenitors (FLB56, FLB99), the fetal liver Pre-B clones (FLB32, FLB41, FLB86), the marrow Pro-T lymphocyte progenitors (C4-77/3, C4-90/16), the fetal thymus Pro-T lymphocyte progenitors (FTH5, FTA2, FTF1), were described and have been propagated in culture as detailed before (Palacios and Samaridis, 1992; Palacios and Steinmetz, 1985; Palacios et al., 1987a, b; Pelkonen et al., 1987). The Pre-T cell line Degos was obtained from Balb/c adult thymus, requires IL2 + IL7 for continuous growth in culture, is CD483TCRαβTCRγδ and has rearranged both alleles of the TCR β and γ genes (R.P. unpublished results). The 97.2 macrophage line arose spontaneously from bone marrow of an X-ray-irradiated AKR/J mouse (R.P. unpublished results). The Cos-1 cells were provided by René Devos (Roche Research, Gent-Belgium). The mature T-cell lines CTLL (Gillis and Smith, 1977) and HT-2 (Watson, 1979) were also used.

Cytokines

Supernatants from X63Ag8 myeloma cells transfected with cDNAs coding for IL2, IL3, IL4, IL5, IL6 (Karasuyama and Melchers, 1988) or from J558L/A2B2/44 myeloma cells transfected with mouse IL7-cDNA (Samaridis et al., 1991), tested for biological activity in proliferative assays as described (Palacios et al., 1987a; Samaridis et al., 1991), were used. The Cos-1/SF5 stable transfectant cells expressing mouse cDNA encoding the membrane form of SF were generated as follows: the XhoI-XhoI 0.85 kb fragment from the SF-cDNA clone, pCAMG, was sub-cloned in the XhoI-site of the expression vector pcDNAI/NEO (In Vitrogen, Lugano, Switzerland) and the resulting clone was designated pcDNAI/Neo SF. Linearized pcDNAI/NeoSF plasmid DNA was introduced into Cos-1 cells by Lipofection using Lipo-fectin Reagent (BRL, Uxbridge, England) as previously described in detail (Samaridis et al., 1991). Stable transfectants were then selected in culture medium (see below) supplemented with G418 (1 mg/ml). They have been propagated and expanded, if required, in the same selection medium indicated above. Cos-1 transfectant cells and their supernatants were tested for their ability to support proliferation of bone-marrow-derived granulocytes essentially as described by Flanagan et al. (1991) with the exception that we have used mitomycin-c-treated Cos-1 transfectant cells in the assays. Transfectant cells that scored positive in the biological assay were expanded and used to prepare total RNA to confirm the expression of the transduced SF-gene by northern blot analy-sis (see below). The Cos-1/SF5 stable transfectant cells produce both membrane-bound and soluble (low levels) SF and were used in the experiments described here.

Antibodies

FITC-, PE-or biotin-conjugated antibodies against Thy1, Lyt2 (CD8), L3T4 (CD4), B-220 (hybridoma 6B2), TCRαβ (hybridoma 57-597), TCRγδ (hybridoma GL3), CD3 (hybridoma 145-2C11), H-2Kk were purchased from Pharmingen (San Diego, California). FITC- and biotin-labelled Mac-1-specific antibody were from Caltag (Zürich, Switzerland). FITC-labelled anti-mouse μ, kappa, lambda Ig chains and PE-streptavidin were from Southern Biotechnology Associates (Birmingham, Alabama). FITC-conjugated anti-rat IgG-specific antibody, mouse IgG and rat IgG were from Jackson Immunoresearch Laboratory. FITC-streptavidin was from Amersham (Zürich, Switzerland). The c-kit-specific antibody ACK2 (Nishikawa et al., 1991) and the Mac-1-specific antibody M1/70 (Springer et al., 1979) were purified by protein-G chromatography (Pharmacia, Uppsala, Sweden). Purified c-kit antibody was biotinylated as described before (Palacios and Leu, 1986).

Cell preparations

Mononuclear cell suspensions from bone marrow, thymus, spleen and fetal liver were prepared free of erythrocytes as described (Palacios et al., 1987a,b), yolk sac mononuclear cells from day 8-8.5 embryos were kindly provided by Beat Imhof (Basel Institute for Immunology). CD4CD8 thymocytes were obtained by two cycles of antibody plus complement killing of total thymocytes as described (Palacios and von Boehmer, 1986). c-kit+ cells from 14-day fetal liver of CBA/Thy1a embryos were isolated by cell sorter using biotin-conjugated c-kit-specific antibody and FITC-strepta-vidin in a FACStar plus instrument (Becton and Dickinson, Mountain View, California) as described in detail before (Palacios and Leu, 1986). Reanalysis of the cell sorter purified cells showed that >98% of the cells were c-kit+. The cells were washed and resus-pended either in culture medium [Iscove’s modified Dulbecco medium supplemented with heat-inactivated fetal calf serum (7.5%), 2-mercaptoethanol (5×10−5 M), L-glutamine (2 mM) and gentamycin (50 μg/ml)] or FACS buffer (PBS pH 7.3, 0.2% BSA, 0.1% sodium azide) as required.

Isolation and analysis of nucleic acids

DNA and total RNA preparation, restriction enzyme digestions, agarose gel electrophoresis, RNA blotting, probe preparations, hybridization procedures and autoradiography were performed as described (Palacios et al., 1989b; Pelkonen et al., 1987; Palacios and Samaridis, 1991; Gutierrez and Palacios, 1991).

DNA probes

The probes used were: IL7 (450 bp SstI-HindIII fragment), SF (850 bp XhoI-XhoI fragment), β-actin (1.1 kb PstI-PstI fragment)isolated DNA fragments and were labelled with 32P by using a random-primer DNA labeling kit (Boehringer Mannheim).

Functional assays

Reconstitution of Scid mice

Cell sorter purified c-kit+ fetal liver cells from day 14 CBA.Thy1a embryos (2 ×105 cells in 0.5 ml PBS) were injected in the lateral vein of the tail of C.B.17 Scid mice that were previously (4-6 hours before injection) exposed to 300 rads of γ-rays. Scid mice that received c-kit+ fetal liver or PBS only (control) were housed in sterile isolators with sterile water and food. Lymphoid repopulation of bone marrow and spleen of these mice was determined by single- and two-color FACS analysis twelve to sixteen weeks after transfer of the c-kit+ fetal liver cells.

Proliferative cell responses

Cos-1/SF5 transfectant cells and wild-type Cos-1 cells were treated with mitomycin C (20 μg/ml at 37°C for 4 hours). Following four washes in 50 ml of PBS each, they were resuspended in culture medium, distributed in flat-bottomed microtiter wells (∼104-2×104 cells/well) and left to adhere to plastic by incubation at 37°C for 2-4 hours. 104 lymphocyte progenitor cells (PR-23, FTH5, CB/Bm7 and FLB41) were added to the microplate wells in the presence or the absence of limited concentrations or rIL2 (3 units/ml), rIL3 (2 units/ml), rIL4 (5 units/ml), rIL7 (20 units/ml), in a final volume of 200 μl of culture medium per well. In some experiments, the c-kit-specific antibody ACK2 or the Mac-1-specific antibody M1/70 (isotype matched control anti-body) were added (final concentrations of 1, 10 and 100 μg per ml) at the beginning of the cultures. Cell proliferation was measured by [3H]thymidine uptake (1 μCi/well) during the last 8 hours of a 48 hour culture period performed at 37°C. The results are expressed as mean counts/minute of triplicate samples per group where the s.e.m. was less than 10.4% of the mean.

These assays were carried out essentially as described before (Palacios et al., 1989a,b; Palacios and Samaridis, 1992) using ET cortical thymic epithelial cells to support T-cell differentiation of the Pro-T cells FTH5 and RP.0.10 bone marrow stromal cells, rIL7 (250 units/ml) and LPS (40 μg/ml) to induce differentiation of the Pre-B cells FLB41 into IgM+ B lymphocytes. FTH5 Pro-T and FLB41 Pre-B cells also were cocultured on monolayers of Cos-1/SF5 transfectant cells or wild-type Cos-1 cells to test whether SF would induce differentiation of c-kit+ lymphocyte pre-cursor clones. All cultures were carried out in six-well costar plates in a final volume of 2 ml of culture medium per well and incubated at 37°C for 6 and 8-10 days. In some experiments, purified c-kit-specific antibody or control Mac-1 antibody were added (final concentrations 10 and 100 μg/ml) at the beginning of the cultures. At the end of the culture period, the haemopoietic cells were harvested, washed and the presence of TCRαβ/CD3+ T-cells and IgM+ B lymphocytes, respectively, was determined by FACS analysis. In all cases, staining of the cells with PgP-1-specific anti-body was also included and served as positive control. The results are expressed as percent TCRαβ/CD3+ T-cells or IgM+ B lymphocytes developed in the cultures.

Expression of SF- and IL7-RNA transcripts in stromal cell lines and thymic epithelial cell lines

Total RNA was isolated from the fetal liver-derived FLS4.1 stromal cells, the bone-marrow-derived RP.0.10 stromal cells, and the thymic epithelial cell lines ET and EA2 as described before (Samaridis et al., 1991; Gutierrez and Palacios, 1991). In experiments addressing the regulation of SF-RNA expression in FLS4.1 stromal cells, the cells were cultured (∼75% confluency) in the presence or the absence of rIL7, rIL4, rIL6 (final concentrations 10, 50, 100, 500 and 1000 units per ml) either single or in combinations at 37°C for 4 and 24 hours. Total RNA was isolated from each experimental group. RNA transcripts from the SF, IL7 and β-actin genes were determined by northern blot analysis as indicated above.

FACS analysis

This was carried out as described in detail previously (Palacios et al. 1989b; Samaridis et al., 1991; Gutierrez and Palacios, 1991). Single- and two-color FACS analysis were performed using a FACScan instrument (Becton and Dickinson, Mountain View, CA). Bone marrow, spleen and thymocytes from CBA/j mice were used as positive controls as required and to set up electronically green and red compensations. Fluorescence emitted by single viable cells was measured with logarithmic amplification. Dead cells were excluded from analysis by forward and side scatter gating. Data collected from 5 ×103-4×104 cells were analyzed with consort 30 software and displayed in the form of fluorescence histograms (single color) or contour plots (two color).

Immunohistochemistry

Embryo sections (4-6 μm) were prepared and acetone-fixed on glass slides and stored at −20°C until use. Immunofluorescence staining of the slides (four previously selected slides per embryo) was carried out using a previously determined optimal concentration of purified c-kit-specific mAb in staining buffer (PBS + 2% BSA + 0.1% sodium azide), slides were then incubated at 20°C for 1 hour. Following two washes with buffer, FITC antirat IgG-specific antibody (final dilution 1:200) was added and the slides were incubated at 4°C for 1 hour. Following three washes with buffer (each for 10 minutes), the slides were mounted under glass cover slips using 50 mM Tris (pH 8.6) containing Gelvatol and analyzed, both under phase-contrast and green fluorescence with an Axiophot Zeiss microscope. In all cases, the entire embryo was scanned for the presence of cells that specifically bound the c-kit antibody. c-kit+ cells usually displayed a ring-dotted pattern of fluorescence, and sometimes a patchy pattern of fluorescence on one pole of the cells was also observed.

c-kit expression on lineage uncommitted and lineage-restricted lymphocyte precursor cell lines

A panel of non-transformed cell lines isolated from bone marrow, liver and thymus of either embryos or young adult mice which represent different stages of lymphocyte development were tested by FACS for c-kit protein expression on the cell membrane. The results showed that the multi-potent progenitor clones PR-5, PR-8 and PR-23 (able to give rise to T-lymphocytes, B-lymphocytes and myeloid cells) are c-kit+. The bipotent B-cell/myelocytic progenitor clones LyD9, LyH7 and LyB9 (able to generate B-lymphocytes and myeloid cells, but not T-lymphocytes), the monopotent Pro-B lymphocyte progenitor clones derived from either adult bone marrow (CB/Bm7, Bc/Bm11) or 14-day fetal liver (FLB56, FLB99) and the Pre-B cell clones isolated from 14-day fetal liver (FLB32, FLB41, FLB86) were all c-kit positive. The monopotent Pro-T lymphocyte progenitor clones obtained from either 14-day fetal thymus (FTH5, FTF1, FTA2) or adult bone marrow (C4-77/3, C4-90/16) were c-kit+ while the late Pre-T cell line Degos isolated from adult thymus and the mature T-cell lines (CTLL, HT-2) were c-kit. These results are illustrated in the form of fluorescence histograms in Fig. 1 and are summarized in Table 1.

Table 1.

c-kit surface expression on cell lines and cell populations representing different stages of lymphocyte development

c-kit surface expression on cell lines and cell populations representing different stages of lymphocyte development
c-kit surface expression on cell lines and cell populations representing different stages of lymphocyte development
Fig. 1.

c-kit protein expression on the cell membrane of non-transformed clones representing different stages of lymphocyte development was assessed by FACS analysis using biotin-conjugated ACK2 c-kit-specific antibody and FITC-streptavidin. Control: samples exposed to rat IgG and FITC-streptavidin.

Fig. 1.

c-kit protein expression on the cell membrane of non-transformed clones representing different stages of lymphocyte development was assessed by FACS analysis using biotin-conjugated ACK2 c-kit-specific antibody and FITC-streptavidin. Control: samples exposed to rat IgG and FITC-streptavidin.

c-kit expression on lymphoid cell populations in the young adult mice

Single-color FACS analysis of lymphoid cells in the bone marrow of young adult C57BL/6 mice showed that 2-5% of these cells express low levels of c-kit and analysis of the myeloid marrow population showed that 11-17.6% of them were c-kit+ (Fig. 2 and Table 1). Two-color FACS analy-sis revealed that most (94-97%) B-220+ B-cell precursors and all IgM+ mature B lymphocytes in the bone marrow did not carry detectable c-kit (Table 1).

Fig. 2.

The presence of c-kit receptor on the cell populations indicated was determined by FACS analysis using biotin-conjugated ACK2 c-Avz-speciiic antibody and FITC-streptavidin. Control: samples exposed to rat IgG and FITC-streptavidin.

Fig. 2.

The presence of c-kit receptor on the cell populations indicated was determined by FACS analysis using biotin-conjugated ACK2 c-Avz-speciiic antibody and FITC-streptavidin. Control: samples exposed to rat IgG and FITC-streptavidin.

Single-color FACS analysis of total thymocytes showed that only 0.2-0.6% of these cells bound the c-kit-specific antibody; two-color FACS analysis revealed that all CD4+/CD8+ thymocytes were c-kit negative and about one seventh of the CD4 CD8 thymocytes (which comprise T-cell precursors) expressed low levels of c-kit (Fig. 2 and Table 1). Finally, all mature T-lymphocytes and B-lymphocytes in the spleen were c-kit negative (Fig. 2 and Table 1).

c-kit expression on cells from yolk sac, liver and thymus in developing C57BL/6 mouse embryos

FACS analysis of mononuclear cells from yolk sac of day 8-8.5 embryos (before fetal blood circulation has started) showed that 10-18% of these cells express low levels of c-kit receptor and that 25-30% mononuclear cells from day-14 fetal liver brightly stained with the c-kit specific anti-body (Fig. 2). Next we searched for c-kit+ cells in the liver and thymus of embryos at different times of development (from day 10 to day 17) by immunofluorescence staining of embryo sections and microscopy analysis. Although c-kit+ cells were observed in various non-haemapoietic tissues (e.g. brain, dermis, gut), we have focussed our analysis to the two main lymphopoietic organs in the developing embryo, namely liver and thymus. c-kit+ haemopoietic cells were readily found in the liver from day 10 of gestation onwards. c-kit+ haemopoietic liver cells increased in both number and fluorescence intensity at day 11 to day 15 and thereafter decreased, such that fewer c-kit+ haemopoietic cells were found in the liver by day 17 of gestation (Fig. 3).

Fig. 3.

c-kit+ cells in the developing liver. The presence of c-kit+ cells in C57BL/6 embryo sections was assessed by immunofluorescence staining and microscopy. Slides in which ACK2 c-kit-specific antibody was omitted in the staining procedure were used as negative controls.

Fig. 3.

c-kit+ cells in the developing liver. The presence of c-kit+ cells in C57BL/6 embryo sections was assessed by immunofluorescence staining and microscopy. Slides in which ACK2 c-kit-specific antibody was omitted in the staining procedure were used as negative controls.

Very few and weakly c-kit+ cells were observed in the thymus primordium at days 10-11 of gestation when the T-cell progenitors have started to colonize this organ. c-kit+ thymic cells increased both in number and fluorescence intensity from day 12 to day 15, but they sharply decreased by day 17 of gestation (Fig. 4). The fluorescence intensity of c-kit-stained cells in the liver was generally higher than that displayed by c-kit positive cells in the thymus (com-pare Figs 3 and 4), probably reflecting higher levels of c-kit protein on the cell membrane of haemopoietic cells in the liver. No positive cells were found in the control samples in which the c-kit-specific antibody was omitted in the staining procedure (Figs 3 and 4). These findings indicate that within the haemopoietic system of the developing mouse embryo, cells bearing c-kit receptors are already present in the yolk sac at day 8 to 8.5 (before fetal blood circulation has started). c-kit+ haemopoietic cells are then found in the liver from day 10 and in the thymus from day 11 of gestation. In these organs, c-kit+ cells expand up to day 15 to 16 and thereafter switch off expression of c-kit as they differentiate further such that by day 17 only few haemopoietic cells are still c-kit+ in these organs.

Fig. 4.

c-kit+ cells in the developing thymus. The presence of c-kit+ cells in C57BL/6 embryo sections was assessed by immunofluorescence staining and microscopy. Slides in which ACK2 c-kit-specific antibody was omitted in the staining procedure were used as negative controls.

Fig. 4.

c-kit+ cells in the developing thymus. The presence of c-kit+ cells in C57BL/6 embryo sections was assessed by immunofluorescence staining and microscopy. Slides in which ACK2 c-kit-specific antibody was omitted in the staining procedure were used as negative controls.

We wished to determine directly whether the c-kit+ haemopoietic cell population in the fetal liver comprises precursors of T- and B-lymphocytes. To this end, we isolated by cell sorter c-kit+ mononuclear cells from day-14 liver of CBAThy1a(H-2k) embryos and tested these cells for their capacity to reconstitute the lymphoid compartments of T- and B-lymphocyte-deficient C.B.17 Scid (H-2d) mice (Bosma et al., 1983). Twelve to sixteen weeks after transfer of c-kit+ fetal liver cells into sublethally irradiated Scid mice, the presence of B-cell precursors (B-220+ IgM), B-lymphocytes (B-220+ IgM+) and T-lymphocytes (CD4+8TCRαβ+, CD48+TCRαβ+) in the bone marrow and/or spleen of these mice were determined by two-color FACS analysis of the lymphoid population (gated by for-ward and side scatters).

Scid mice that received c-kit+ fetal liver mononuclear cells had reconstituted the B-cell precursor and the mature B-lymphocyte populations in the bone marrow. Most, if not all, of the lymphoid populations in the bone marrow of these mice were H-2k+ indicating that they originated from the donor c-kit+ fetal liver cells (CBAThy1a, H-2k+). Few, if any, cells in this population stained with the myeloid sur-face marker Mac-1 further documenting the lymphoid nature of the cells examined (Fig. 5). Analysis of the lym-phoid population in the spleens of Scid mice that received c-kit+ fetal liver cells indicated that both T- and B-lym-phocyte compartments were reconstituted in these animals. Firstly, most (>93%) lymphoid cells were H-2k+ which shows that they were the progeny of the c-kit+ fetal liver precursor cells injected (CBAThy1a origin, H-2k) and not of host origin (C.B.17, H-2d). Secondly, both major sub-sets of mature T-lymphocytes(CD4+8TCRαβ+,CD48+TCRαβ+) and mature B-lymphocytes (B-220+ IgM+) expressing normal ratios of kappa and lambda Ig light chains were present in the spleens of the reconstituted Scid mice (Fig. 5). Details of the results obtained in another four scid mice reconstituted with purified c-kit+ fetal liver mononuclear cells and in three control scid mice that received no cells are given in Table 2. We conclude that c-kit+ fetal liver mononuclear cells comprise precursors of both T- and B-lymphocytes.

Table 2.

c-kit+14-day fetal liver cells contain precursors of both T- and B-lymphocytes

c-kit+14-day fetal liver cells contain precursors of both T- and B-lymphocytes
c-kit+14-day fetal liver cells contain precursors of both T- and B-lymphocytes
Fig. 5.

Cell sorter purified c-kit+ 14-day fetal liver cells reconstitute the lymphoid compartments of C.B. 17 Scid mice. The presence of the lymphocyte populations in the bone marrow and spleen of Scid mice that had received c-kit+ purified fetal liver cells sixteen weeks before was determined by FACS analysis of the “lymphoid” population (gated by forward and side scatters).

Fig. 5.

Cell sorter purified c-kit+ 14-day fetal liver cells reconstitute the lymphoid compartments of C.B. 17 Scid mice. The presence of the lymphocyte populations in the bone marrow and spleen of Scid mice that had received c-kit+ purified fetal liver cells sixteen weeks before was determined by FACS analysis of the “lymphoid” population (gated by forward and side scatters).

Stromal cell lines from fetal liver and bone marrow and thymic epithelial cell lines synthesize mRNA for SF

We considered it important to determine whether stromal cells from fetal liver and bone marrow of young adult mice and thymic epithelial cell lines, previously shown to support growth/differentiation of B-or T-lymphocyte progenitors (Palacios et al., 1989a,b; Gutierrez and Palacios, 1991; Palacios and Samaridis, 1992) synthesize the ligand for c-kit. Northern blot analysis showed that the FLS4.1 stromal line derived from 15-day fetal liver and the RP.0.10 stromal line isolated from adult bone marrow synthesize RNA transcripts for SF. FLS4.1 cells produce higher levels of SF-RNA than RP.0.10 cells (Fig. 6A). The cortical thymic epithelial lines EA2 (isolated from newborn thymus) and ET (derived from 16-day fetal thymus), but not the bone marrow-derived macrophage line 97-2, synthesize SF-RNA (Fig. 6A). All RNA samples hybridized to the actin probe (Fig. 6A). Thus, stromal cells and thymic epithelial cells with known capacity to support growth/differentiation of c-kit+ lymphocyte progenitors, produce SF.

Fig. 6.

(A) Northern blot analysis of total RNA (10 μg/lane) from the fetal liver derived FLS4.1 stromal cells, the adult bone marrow-derived RP.0.10 stromal cells, the ET and EA2 cortical thymic epithelial cells and the bone-marrow-derived 97.2 macrophage cell line, to assess the expression of SF, IL7 and β-actin genes. The results shown above are from films exposed for 18 hours (SF and IL7 probes) or for 6 hours (actin probe). (B) Cytokine-regulation of SF-RNA synthesis in FLS4.1 stromal cells. Northern blot analysis of total RNA (10 μg/lane) from FLS4.1 stromal cells which were either cultured in medium or exposed to cytokines for 4 hours (see Materials and methods for details). Lane A: FLS4.1 cells cultured in medium only. Lanes B to F: cells exposed to 10, 50, 100, 500 and 1000 units/ml of rIL7, respectively. Lanes G and H: cells exposed to rIL7 (500 units/ml) plus rIL4 (10 and 100 units/ml, respectively). Lanes I and J: cells exposed to rIL7 (500 units/ml) plus rIL6 (10 and 100 units/ml, respectively). The results shown above are from films exposed for 18 hours (SF probe) and 6 hours (actin probe). (C) The expression of the SF-RNA transcripts in Cos-1/SF5 cells transfected with the pcDNAI/NeoSF plasmid and in wild type Cos-1 cells was assessed by Northern blot analysis of total RNA (10 μg/lane). Films were exposed as indicated in B.

Fig. 6.

(A) Northern blot analysis of total RNA (10 μg/lane) from the fetal liver derived FLS4.1 stromal cells, the adult bone marrow-derived RP.0.10 stromal cells, the ET and EA2 cortical thymic epithelial cells and the bone-marrow-derived 97.2 macrophage cell line, to assess the expression of SF, IL7 and β-actin genes. The results shown above are from films exposed for 18 hours (SF and IL7 probes) or for 6 hours (actin probe). (B) Cytokine-regulation of SF-RNA synthesis in FLS4.1 stromal cells. Northern blot analysis of total RNA (10 μg/lane) from FLS4.1 stromal cells which were either cultured in medium or exposed to cytokines for 4 hours (see Materials and methods for details). Lane A: FLS4.1 cells cultured in medium only. Lanes B to F: cells exposed to 10, 50, 100, 500 and 1000 units/ml of rIL7, respectively. Lanes G and H: cells exposed to rIL7 (500 units/ml) plus rIL4 (10 and 100 units/ml, respectively). Lanes I and J: cells exposed to rIL7 (500 units/ml) plus rIL6 (10 and 100 units/ml, respectively). The results shown above are from films exposed for 18 hours (SF probe) and 6 hours (actin probe). (C) The expression of the SF-RNA transcripts in Cos-1/SF5 cells transfected with the pcDNAI/NeoSF plasmid and in wild type Cos-1 cells was assessed by Northern blot analysis of total RNA (10 μg/lane). Films were exposed as indicated in B.

rIL7 down regulates the synthesis of SF-RNA in stromal cells and rIL6 abrogates the inhibitory effect of IL7

Intriguingly, the stromal cells and thymic epithelial cells that produced higher levels of SF-mRNA (FLS4.1, EA2) synthesized much less RNA transcripts from the IL7 gene while stromal cells and thymic epithelial cells that produced low levels of SF-RNA transcripts (RP.0.10, ET) synthesized higher levels of IL7-RNA (Fig. 6A). This reciprocal correlation suggested that IL7 and SF could negatively regulate each other’s production. We studied this possibility in FLS4.1 stromal cells as they constitutively expressed the highest levels of SF-RNA but not detectable RNA transcripts for IL1 to IL7 (Fig. 6A, Palacios and Samaridis, 1992). We found that exposure of FLS4.1 cells to rIL7 dramatically decreased the steady-state RNA synthesis for SF as determined by northern blot analysis. The inhibitory effect of rIL7 was observed in FLS4.1 cells exposed to this cytokine for 4 and 24 hours and it was apparent at rIL7 concentrations of 10 units per ml or higher. We also tested two other cytokines often produced by stromal cells (among other cell types), rIL4 and rIL6. Although neither rIL4 nor rIL6 had by itself significant effects on the synthesis of SF-RNA in FLS4.1 stromal cells, rIL6, but not rIL4, abrogated the inhibitory effect of rIL7 on SF-RNA synthesis in these cells (Fig. 6B and data not shown). All samples hybridized equally to the actin probe (Fig. 6B). These results provide evidence that the synthesis of SF-RNA in stromal cells is subject to regulation by cytokines that can be produced by the same cells or by neighboring cells. The data also point out that the regulation of the SF expression is complex and that the inhibitory effect of one cytokine (i.e. IL7) can be abolished by another one (i.e. IL6) even in the continuous presence of the inhibitory cytokine.

Recombinant membrane-bound SF supports survival and enhances the proliferative responses of c-kit+ lymphocyte progenitor clones to other growth factors

The next set of experiments aimed to assess the effects of c-kit/SF on growth and/or differentiation of lymphocyte progenitor clones. To address this directly, we first generated stable Cos-1 cell transfectants expressing the membrane form of mouse SF (see Materials and methods for details). Northern blot analysis confirmed that Cos-1/SF5 transfectants synthesize SF while wild-type Cos-1 cells do not (Fig. 6C).

Coculture of c-kit+ progenitor clones representing various stages of T- and B-lymphocyte development (multipo-tent progenitor PR-23, Pro-T cells FTH5, Pro-B cells CB/Bm7 and Pre-B cells FLB41) with Cos-1/SF5 trans-fectants showed that all progenitor cells survived for 6 to 10 days. By contrast, the same progenitor cells died within 2 days when cultured on monolayers of wild-type Cos-1 cells or in culture medium lacking their required exogenous growth factors. Cos-1/SF5 transfectants did not by them-selves support significant proliferation of any of the c-kit+ lymphocyte progenitor clones tested, but they did significantly enhance the proliferative response of the lymphoid progenitor clones to suboptimal concentrations of their respective growth factors, rIL2, rIL3, rIL4 or rIL7. The latter effect was not observed with the wild-type Cos-1 cells that do not synthesize SF (Table 3). The enhancing effect of Cos-1/SF5 tranfectants on the proliferative responses to rIL2, rIL3, rIL4 or rIL7 was inhibited by a c-kit-specific antibody but not by an isotype-matched irrelevant control antibody (Fig. 7). These results indicate that the interaction of SF on Cos-1/SF5 transfectant cells with c-kit receptors on the lymphocyte progenitors was responsible for the growth-enhancing effects observed.

Table 3.

rSF expressed on Cos-1 cells enhances the proliferative response of lymphocyte progenitor clones to interleukins 2, 3, 4 and 7

rSF expressed on Cos-1 cells enhances the proliferative response of lymphocyte progenitor clones to interleukins 2, 3, 4 and 7
rSF expressed on Cos-1 cells enhances the proliferative response of lymphocyte progenitor clones to interleukins 2, 3, 4 and 7
Fig. 7.

c-kit antibody inhibits the enhanced proliferative responses of lymphocyte progenitor clones FTH5 to rIL2 (A), CB/Bm7 to rIL3 (B), and FLB41 to rIL7 (C), supported by SF-bearing Cos-1/SF5 transfectant cells, but it does not affect the response of these cells to the same suboptimal concentrations of growth factors in cultures carried out without Cos-1/SF5 transfectant cells. See Materials and methods for details. The results are the mean ± s.e.m. of triplicate samples per group.

Fig. 7.

c-kit antibody inhibits the enhanced proliferative responses of lymphocyte progenitor clones FTH5 to rIL2 (A), CB/Bm7 to rIL3 (B), and FLB41 to rIL7 (C), supported by SF-bearing Cos-1/SF5 transfectant cells, but it does not affect the response of these cells to the same suboptimal concentrations of growth factors in cultures carried out without Cos-1/SF5 transfectant cells. See Materials and methods for details. The results are the mean ± s.e.m. of triplicate samples per group.

In contrast to the effects of SF on cell survival and growth of c-kit+ lymphocyte progenitors observed, we did not find evidence that Cos-1/SF5 transfectant cells could by themselves support differentiation of c-kit+ T-or B-lymphocyte progenitor clones into mature T- and B-lymphocytes, respectively, in in vitro coculture assays (Table 4). Nor did we find any evidence that the c-kit-specific anti-body affected the differentiation of Pro-T cells into mature TCRαβ/CD3+ T-cells supported by the ET cortical thymic epithelial cells or that of Pre-B cells into IgM+ B lymphocytes supported by the RP.0.10 stromal cells, rIL7 and LPS (Table 4). These results argue that the c-kit/SF system does not play an essential role in the differentiation of lymphocyte progenitors into mature T-or B-lymphocytes. The c-kit receptor/SF ligand pair seems to participate in the development of the lymphoid system by supporting cell survival and by enhancing the proliferation of lymphocyte progenitors to other growth factors (e.g. IL2, IL3, IL4, IL7).

Table 4.

The c-kit/SF is neither sufficient nor essential for differentiation of lymphocyte progenitor clones into mature T-or B-lymphocytes

The c-kit/SF is neither sufficient nor essential for differentiation of lymphocyte progenitor clones into mature T-or B-lymphocytes
The c-kit/SF is neither sufficient nor essential for differentiation of lymphocyte progenitor clones into mature T-or B-lymphocytes

Considerable information has been obtained on the expression and function of the c-kit receptor/SF ligand pair in germ cells, nerve cells, melanoblasts and erythroid and myeloid cell lineages of the haemopoietic system in the last few years (Orr-Urtreger et al., 1990; Matsui et al., 1990; Nishikawa et al., 1991; Ogawa et al., 1991; Dolci et al., 1991; Keshet et al., 1991; Migliaccio et al., 1991; McNiece et al., 1991a,b). Here we describe our studies on the expression and participation of the c-kit/SF system during lymphocyte development both in the embryo and in young adult normal mice. Our results obtained with lymphocyte progenitor clones isolated from liver or thymus of embryos, with freshly isolated haemopoietic cells from yolk sac and liver and with embryo sections, strongly suggest that the c-kit/SF system plays an important role in early stages of development of lymphocyte precursors in the mouse embryo. This conclusion stems from the findings that: (a) Pro-T, Pro-B and Pre-B lymphocyte precursor clones derived from thymus or liver of 14-day embryos express functional c-kit receptors on the cell membrane, (b) c-kit+ cells are quite well represented in tissues known to contain lymphocyte progenitors, from day 8-8.5 in yolk sac, day 10 in liver, day 11-12 in thymus of the developing embryo. The number of c-kit+ cells in the liver and the thymus increases up to day 15 and thereafter progressively decreases, (c) purified c-kit+ day 14 fetal liver cells fully reconstitute the T- and B-lymphocyte compartments of Scid mice, and (d) fetal liver-derived stromal cells and fetal thymus-derived thymic epithelial cells able to support growth/differentiation of c-kit+ lymphocyte progenitor clones, synthesize RNA transcripts for SF, the ligand for c-kit receptor. Moreover, very recent work by other investigators has documented the synthesis of SF-mRNA in yolk sac, liver and thymus of the developing embryo (Matsui et al., 1990; Keshet et al., 1991; Wiles et al., 1992).

Unlike the findings obtained in the developing embryo, our studies with lymphocyte progenitor clones and with freshly isolated lymphocyte precursors and mature lymphocytes from bone marrow, thymus and spleen of adult normal mice indicate that the cell surface expression of c-kit receptor is restricted to cells at very early stages of lymphocyte development (i.e. multipotent progenitors, B-cell/myelocytic bipotent progenitors, monopotent Pro-T and Pro-B lymphocyte progenitors). A recent study shows that pluripotent haemopoietic stem cells in the bone marrow of adult mice express c-kit receptor (Ikuta and Weissman, 1992). The best examples that illustrate this difference between embryo and adult mice are that: (a) fetal liver B-cell precursors are c-kit+ while most B-cell precursors in the adult bone marrow are c-kit and (b) most immature and mature thymocytes in the adult mice are c-kit whereas a considerable proportion of thymocytes in the fetus are c-kit+. The findings in the adult mouse lymphoid compart-ments point out that, although the c-kit/SF system may con-tribute to the formation of very early lymphocyte progenitors, this receptor/ligand pair no longer participates in later stages of lymphocyte development (from Pre-T or Pre-B stages onwards). Other receptor/growth factor pairs (e.g. IL7, IL4, IL3, IL2) must take over/replace the c-kit/SF system as the Pro-T and Pro-B cell precursors develop in the adult mice. This view is strengthened by and explains the finding that treatment of mice with the c-kit-specific antibody ACK2 did not affect development of lymphocytes while it abolished the generation of myeloid and erythroid-precursor cells (Ogawa et al., 1991). Furthermore, the fact that mice having less severe forms of c-kit/SF deficiency which reach adulthood are not lymphocyte deficient (Russel 1979) is also consistent with such scenario. It will be now important to identify molecules that regulate the expression of c-kit on lymphoid cells from embryo and adult mice.

Our findings in the embryo and adult mice described here add further evidence to the existence of differences in functions of both haemopoietic precursor cells and microenvironments for lymphocyte differentiation between the embryo and the adult mouse (Ikuta et al., 1990; Palacios et al., 1989a).

As to the actual functions of the c-kit/SF system in the development of lymphocyte precursors, our results indicate that c-kit/SF may contribute to this process by supporting cell survival and by enhancing the response of the developing lymphoid precursor to other growth factors (e.g. IL2, IL3, IL4, IL7). Similar results were obtained with B cell precursors freshly isolated from 14 day fetal liver and bone marrow of adult mice (Palacios and Samaridis, 1992; McNiece et al., 1991b). These effects are not peculiar to the lymphocyte precursor population as c-kit/SF also exerts similar functions on primordial germ cells, melanoblasts and myeloid-lineage haemopoietic cells (Dolci et al., 1991; Lowry et al., 1991; McNiece et al., 1991a). Thus, cell survival and enhancement of cell proliferation supported by other cytokines seems a general function of the c-kit/SF system.

Another finding of fundamental interest in the present study is that the synthesis of SF-RNA transcripts in stromal cells is subject to regulation by cytokines that can be produced by the same cells or neighboring cells. Moreover, the regulatory effect of a given cytokine on SF-RNA synthesis is itself susceptible to regulation by yet another cytokine. This complex regulatory network is nicely illustrated by the finding that rIL7 down regulates SF-RNA synthesis in FLS4.1 stromal cells and that rIL6 abrogates this negative effect of rIL7. That such regulatory events found in FLS4.1 stromal cells may also apply to other stromal cells and thymic epithelial cells is likely because stromal cells and thymic epithelial cells that synthesize higher levels of IL7 make little SF-RNA while cells that synthesize higher levels of SF-mRNA produce no or low levels of IL7-RNA transcripts. This led us to the view that the expression and the levels of, at least, SF and IL7 produced by a given stromal or thymic epithelial cell are not only a reflection of the inherent potential of the cell but also the result of positive and negative regulation by cytokines that can be produced by the same cell or by neighboring cells in particular microenvironments at a given time in ontogeny. It will be now important to determine the molecular mechanisms by which IL7 down regulates expression of SF-mRNA and that by which IL6 abrogates this negative effect of IL7 on stromal cells. Experiments to address these issues are now underway in our laboratories.

In summary, our results described here can be integrated in the following scenario concerning the participation of c-kit/SF in the development of lymphocytes. c-kit+ lymphocyte progenitors interact with SF-bearing stromal cells in the fetal liver and bone marrow or epithelial cells in the thymus. This event is likely facilitated by adhesion molecules (Imhof et al., 1991; Miyake et al., 1991), although there is evidence that c-kit/SF could act in this fashion too (Flanagan et al., 1991). The result of this c-kit/SF interaction is survival and enhanced proliferative response of the progenitor cells to other growth factors as they become available. Here, c-kit/SF not only could render the lymphoid progenitor cells more sensitive to the action of other growth factors (i.e. lower threshold to trigger proliferation) but it also could facilitate the exposure of these cells to adequate levels of appropriate combinations of other cytokines. It is also possible that binding of the c-kit receptor on lymphocyte progenitors to SF-bearing stromal or thymic epithelial cells also results in stimulation or inhibition of the synthesis of cytokines either in the stromal and thymic epithelial cells, in the developing haemopoietic precursor itself or in both. The proliferating lymphocyte progenitor attached to the stromal cells and/or the thymic epithelial cells are induced by the latter cells to express their genetic program. The precise mechanisms by which stromal cells or thymic epithelial cells induce differentiation of lymphocyte precursors into mature lymphocytes are still undefined. Our results described here argue that the c-kit/SF system does not play a mandatory role in this process.

We are grateful to J. Samaridis for excellent technical assis-tance, H. Karasuyama and S. Carson for cytokines and actin probe, B. Imhof for yolk sac cell preparations, D. Thorpe for help with the cell sorter, B. Collins and E. Wagner for animal care, W. Haas, H. von Boehmer and F. Ronchese for critical reading of the man-uscript, H. P. Stahlberger for preparation of the illustrations and N. Schoepflin for excellent preparation of the manuscript. The Basel Institute for Immunology was founded and is supported by F. Hoffmann-La Roche Ltd., Basel, Switzerland.

Bosma
,
G.
,
Custer
,
R.
and
Bosma
,
M. J.
(
1983
).
A severe combined immunodeficiency mutation in the mouse
.
Nature
301
,
527
529
.
Broxmeyer
,
H.
,
Hangol
,
G.
,
Cooper
,
S.
,
Anderson
,
D.
,
Cosman
,
D.
,
Lyman
,
S.
and
Williams
,
D. E.
(
1991
).
Influence of murine mast cell growth factor (c-kit ligand) on colony formation of mouse marrow hematopoietic progenitor cells
.
Exp. Haematol
.
19
,
143
146
.
Chabot
,
B.
,
Stephenson
,
D.
,
Chapman
,
V.
,
Besmer
,
P.
and
Bernstein
,
A.
(
1988
).
The protooncogene c-kit encoding a transmembrane tyrosine kinase receptor maps to the mouse W locus
.
Nature
335
,
88
90
.
Dolci
,
S.
,
Williams
,
D. E.
,
Ernst
,
M.
,
Resnick
,
J.
,
Brannan
,
C.
,
Lock
,
L.
,
Lyman
,
S.
,
Scott-Boswell
,
H.
and
Donovan
,
P.
(
1991
).
Requirement for mast cell growth factor for primordial germ cell survival in culture
.
Nature
352
,
809
811
.
Flanagan
,
J.
,
Chan
,
D.
and
Leder
,
P.
(
1991
).
Transmembrane form of the kit-ligand growth factor is determined by alternative splicing and is missing in the Sld mutant
.
Cell
64
,
1025
1035
.
Geissler
,
E.
,
Ryan
,
M
and
Housman
,
D. E.
(
1988
).
The dominant white spotting (w) locus of the mouse encodes the c-kit proto-oncogene
.
Cell
55
,
185
192
.
Gillis
,
S.
and
Smith
,
K. A.
(
1977
).
Long term culture of tumor specific cytotoxic T cells
.
Nature
262
,
154
156
.
Gutierrez
,
J. C.
and
Palacios
,
R.
(
1991
).
Heterogeneity of thymic epithelial cells in promoting T-lymphocyte differentiation in vivo
.
Proc. Natl. Acad. Sci. USA
88
,
642
646
.
Huang
,
E.
,
Nocka
,
K.
,
Beier
,
D.
,
Chu
,
T.
,
Buck
,
J.
,
Lahm
,
H.
,
Wellner
,
D.
,
Leder
,
P.
and
Besmer
,
P.
(
1990
).
The hematopoietic growth factor KL is encoded by the Sl locus and is the ligand of the c-kit receptor, the gene product of the w locus
.
Cell
63
,
225
233
.
Ikuta
,
K.
,
Kina
,
T.
,
MacNeil
,
I.
,
Uchida
,
N.
,
Peult
,
B.
,
Chien
,
Y.-H.
and
Weissman
,
I. L.
(
1990
).
A developmental switch in thymic lymphocyte maturation potential occurs at the level of hematopoietic stem cells
.
Cell
62
,
836
874
.
Ikuta
,
K.
and
Weissman
,
I. L.
(
1992
).
Evidence that the hematopoietic stem cells express mouse c-kit but do not depend on Steel factor for their generation
.
Proc. Natl. Acad. Sci. USA
89
,
1502
1506
.
Imhof
,
B.
,
Ruiz
,
P.
,
Hesse
,
B.
,
Palacios
,
R.
and
Dunon
,
D.
(
1991
).
EA-1, a novel adhesion molecule involved in the homing of progenitor T lymphocytes to the thymus
.
J. Cell Biol
.
114
,
1069
1078
.
Karasuyama
,
H.
and
Melchers
,
F.
(
1988
).
Establishment of mouse cell lines which constitutively secrete large quantities of interleukin 2, 3, 4 and 5 using modified cDNA expression vectors
.
Eur. J. Immunol
.
18
,
97
106
.
Keshet
,
E.
,
Lyman
,
S.
,
Williams
,
D.
,
Anderson
,
D.
,
Jenkins
,
N.
,
Copeland
,
N.
and
Parada
,
L.
(
1991
).
Embryonic RNA expression patterns of the c-kit receptor and its cognate ligand suggest multiple functional roles in mouse development
.
EMBO J
.
10
,
2425
2435
.
Lowry
,
P.
,
Zsebo
,
K.
,
Deacon
,
D.
,
Eichman
,
C.
and
Quesenberry
,
P. J.
(
1991
).
Effects of rrSCF on multiple cytokine response HPP-CFC generated from Sca1+ lin murine hematopoietic progenitors
.
Exp. Hematol
.
19
,
994
996
.
Matsui
,
Y.
,
Zsebo
,
K.
and
Hogan
,
B.
(
1990
).
Embryonic expression of a hematopoietic growth factor encoded by the Sl locus, the ligand for c-kit
.
Nature
347
,
667
669
.
McNiece
,
I. K.
,
Langley
,
K. E.
and
Zsebo
,
K.
(
1991a
).
Recombinant human stem cell factor synergizes with GM-CSF, G-CSF, IL3, EPo to stimulate human progenitor cells of the myeloid and erythroid lineages
.
Exp. Haematol
.
19
,
226
231
.
McNiece
,
I. K.
,
Langley
,
K. E.
and
Zsebo
,
K.
(
1991b
).
The role of stem cell factor in early B-cell development. Synergistic interaction with IL7
.
J. Immunol
.
146
,
3785
3790
.
Migliaccio
,
G.
,
Migliaccio
,
R.
,
Valinsky
,
J.
,
Longley
,
K.
,
Zsebo
,
K.
,
Visser
,
J. W. M.
and
Adamson
,
J. W.
(
1991
).
Stem cell factor induces proliferation and differentiation of highly enriched murine haematopoietic cells
.
Proc. Natl. Acad. Sci. USA
88
,
7420
7424
.
Miyake
,
K.
,
Weissman
,
I. L.
,
Greenberger
,
J. S.
and
Kincade
,
P.
(
1991
).
Evidence for a role of the integrin VLA-4 in lympho-hemopoiesis
.
J. Exp. Med
.
173
,
599
607
.
Nishikawa
,
S.
,
Kusakabo
,
M.
,
Yoshinaga
,
K.
,
Ogawa
,
M.
,
Hayahsi
,
S.
,
Kunisada
,
T.
,
Era
,
T.
,
Sakakura
,
T.
and
Nishikawa
,
S. I.
(
1991
).
In utero manipulation of coat color formation by a monoclonal anti-c-kit antibody: two distinct waves of c-kit dependency during melanocyte development
.
EMBO J
.
10
,
2111
2118
.
Ogawa
,
M.
,
Matsutaki
,
Y.
,
Sudo
,
T.
,
Kira
,
T.
,
Nakauchi
,
H.
and
Nishikawa
,
S. I.
(
1991
).
Expression and function of c-kit in hemopoietic progenitor cells
.
J. Exp. Med
.
174
,
63
71
.
Orr-Urtreger
,
A.
,
Avivi
,
A.
,
Zimmer
,
Y.
,
Givol
,
D.
,
Yarden
,
Y.
and
Lonai
,
P.
(
1990
),
A proto-oncogene encoded by the W locus
.
Development
109
,
911
923
.
Palacios
,
R.
and
Steinmetz
,
M.
(
1985
).
IL3-dependent mouse clones that express B-220 surface antigen, contain genes in germ-line configuration and generate B lymphocytes in vivo
.
Cell
41
,
727
734
.
Palacios
,
R.
and
von Boehmer
,
H.
(
1986
).
Requirements for growth in vitro of immature thymocytes from adult and fetal mice
.
Eur. J. Immunol
.
16
,
12
18
.
Palacios
,
R.
and
Leu
,
T.
(
1986
).
CC11: a monoclonal antibody specific for Interleukin 3-sensitive mouse cells defines two major populations of B cell precursors in the bone marrow of young adult mice
.
Immunol. Rev
.
93
,
125
145
.
Palacios
,
R.
,
Karasuyama
,
H.
and
Rolink
,
A.
(
1987a
).
Ly1+ pro-B lymphocyte clones. Phenotype, growth requirements and differentiation in vitro and in vivo
.
EMBO J
.
6
,
3687
3693
.
Palacios
,
R.
,
Kiefer
,
M.
,
Brockhaus
,
M.
,
Karjalainen
,
K.
,
Dembic
,
Z.
,
Kisielow
,
P.
and
von Boehmer
,
H.
(
1987b
).
Molecular, cellular and functional properties of bone marrow T lymphocyte progenitor clones
.
J. Exp. Med
.
166
,
12
27
.
Palacios
,
R.
,
Studer
,
S.
and
Rolink
,
A.
(
1989a
).
The epigenetic influence of fetal liver and bone marrow stroma on the developmental potential of Ly1+ Pro-B lymphocyte clones
.
Eur. J. Immunol
.
19
,
347
356
.
Palacios
,
R.
,
Studer
,
S.
,
Samaridis
,
J.
and
Pelkonen
,
J.
(
1989b
).
Thymic epithelial cells induce in vitro differentiation of Pro-T lymphocyte clones into TCRαβ/T3+ and TCRγδ/T3+ cells
.
EMBO J
.
8
,
4053
4063
.
Palacios
,
R.
and
Samaridis
,
J.
(
1992
).
Fetal liver Pro-B and Pre-B lymphocyte clones. Expression of lymphoid specific genes, growth requirements, colonization of the bone marrow and generation of B lymphocytes in vivo and in vitro
.
Mol. Cell. Biol
.
12
,
518
530
.
Pelkonen
,
J.
,
Sideras
,
P.
,
Ramensee
,
H. G.
,
Karjalainen
,
K.
and
Palacios
,
R.
(
1987
).
Thymocyte clones from 14-day mouse embryos I. State of T cell receptor genes, surface markers, and growth requirements. J.Exp. Med
.
166
,
1245
1258
.
Russel
,
E. S.
(
1979
).
Hereditary anemias of the mouse: a review for geneticists
.
Adv. Genet
.
20
,
357
459
.
Samaridis
,
J.
,
Casorati
,
G.
,
Traunecker
,
A.
,
Iglesias
,
A.
,
Gutierrez
,
J. C.
,
Müller
,
U.
and
Palacios
,
R.
(
1991
).
Development of lymphocytes in interleukin-7 transgenic mice
.
Eur. J. Immunol
.
21
,
453
460
.
Springer
,
T.
,
Galfre
,
G.
,
Secher
,
D. S.
and
Milstein
,
C.
(
1979
).
Mac-1: a macrophage differentiation antigen identified by monoclonal antibody
.
Eur. J. Immunol
.
9
,
301
306
.
Watson
.
J.
(
1979
).
Continuous proliferation of murine antigen-specific helper T lymphocytes in culture
.
J. Exp. Med
.
150
,
1510
1521
.
Wiles
,
M.
,
Ruiz
,
P.
and
Imhof
,
B. A.
(
1992
).
Interleukin-7 expression during mouse thymus development
.
Eur. J. Immunol
.
22
,
1037
1042
.
Williams
,
D. E.
,
Eisenman
,
J.
,
Baird
,
A.
,
Rauch
,
C.
,
Van Ness
,
K.
,
March
,
C. J.
,
Park
,
L. S.
,
Martin
,
U.
,
Mochizuki
,
D. Y.
,
Boswell
,
H. S.
,
Burgess
,
G. S.
,
Cosman
,
D.
and
Lyman
,
S.
(
1990
).
Identification of a ligand for the c-kit proto-oncogene
.
Cell
63
,
167
174
.
Witte
,
O. N.
(
1990
).
Steel locus define new multipotent growth factor
.
Cell
63
,
5
6
.
Zsebo
,
K. M.
,
Williams
,
D. A.
,
Geissler
,
E. N.
,
Broudy
,
V. C.
,
Martin
,
F. H.
,
Atkins
,
H. L.
,
Hsu
,
R. Y.
,
Birkett
,
N. C.
,
Okino
,
K. H.
,
Murdock
,
D. C.
,
Jacobsen
,
F. W.
,
Langley
,
K. E.
,
Smith
,
K. A.
,
Takeishi
,
T.
,
Cattanach
,
B. M.
,
Galli
,
S. J.
and
Suggs
,
S. V.
(
1990
).
Stem cell factor is encoded at the Sl locus of the mouse and is the ligand for the c-kit tyrosine kinase receptor
.
Cell
63
,
213
224
.