|
|||||||||
|
|
|
|
|
|
|
|
|
|
|
HEMATOPOIESIS
From the Unité du Développement des
Lymphocytes, and Unité des Cytokines et
Développement Lymphoïde, Département
d'Immunologie, Institut Pasteur, Paris, France.
This article describes the isolation of a novel cell population
(B220loc-kit+CD19 Multipotent hematopoietic stem cells (HSCs)
differentiate into precursors with increasingly restricted
differentiation potential. This process, called lineage commitment,
leads to the generation of oligopotent progenitors and finally to cells
that are irreversibly engaged in a unique pathway of differentiation.
The identification of progenitors at intermediate stages of
differentiation is a fundamental step in understanding lineage
commitment. The characterization of such intermediates within the
erythromyeloid lineages has identified, in the bone marrow (BM), a
granulocyte-macrophage-restricted precursor and, more recently, a
common myeloid precursor and an erythrocyte-megakaryocyte precursor.1
T-cell generation occurs in the thymus from hematopoietic precursors
present in fetal liver (FL) and adult BM. Although the pathways of
intrathymic T-cell differentiation are relatively well characterized,
the role of the thymic microenvironment as a unique site for the
induction of commitment to the T-cell lineage remains a matter of
debate.2,3 Moreover, the identification of T-cell
progenitors before thymic colonization remains incomplete. Cells
endowed with a nonrestricted potential of differentiation into T
lymphocytes have been phenotypically characterized and isolated from
adult BM. Thus, HSCs as well as common lymphoid progenitors (CLPs) were
purified.4,5
The existence of committed T-cell precursors (TCPs) in the
BM6 and more recently in the FL has been
suggested.7,8 However, the identification of surface
markers allowing the isolation of these cells has not been achieved.
Populations of precursors restricted to the T-cell and natural killer
cell (NK) lineages have been reported in fetal thymus, blood,
and spleen.9-13 In 2 independent studies, T and NK
precursors have been characterized in fetal thymus either as
Fc We and others have previously reported that the FL, the major
hematopoietic organ during embryonic life, provides TCPs with restricted potential of differentiation that continuously colonize the
fetal thymus.14-16 In an attempt to identify such
intermediate precursors, we isolated different fractions of FL cells
and analyzed them in a quantitative manner for lymphocyte precursor activity.
Here we identify a novel population corresponding to 0.2% of FL cells
that includes the majority (70%) of TCPs in this organ. These cells
retained the capacity to differentiate into both T and NK progeny at
the single-cell level. When transferred in vivo, they reconstituted the
peripheral T and NK compartments and gave rise to intraepithelial T
cells of extrathymic origin. This cell population, designated
here as common T/NK cell progenitor (C-TNKP), differs both by
surface marker and by gene expression analysis from previously
described bipotent T/NK precursors present in fetal blood, spleen, and
thymus. We propose that these cells represent the immediate
developmental step before thymic immigration.
Mice
Cell preparation, immunofluorescence staining, and cell
sorting
In vitro assays TCP potential.
FL cell suspensions were prepared from 15 to 20 embryos (B6-Ly5.1). TCP
potential was determined using the fetal thymic organ culture (FTOC)
assay,18 with a slight modification.15 Thymic lobes from day-14 fetuses (B6-Ly5.2) were irradiated with 3000 rad
using a cesium B-cell precursor potential. The limiting dilution culture conditions for B-cell precursor potential were described previously.19 After 10 to 14 days of culture, developing pre-B-cell colonies were scored on an inverted microscope. Individual representative colonies were further analyzed by flow cytometry. Cells from each well were further stimulated with lipopolysaccharide (LPS; Salmonella typhosa WO901; Difco) at a final concentration of 25 µg/mL, and IgM secretion was detected by enzyme-linked immunosorbent assay. Frequencies of B-cell precursors were determined according to the Poisson distribution. NK cell precursor potential. NK cell precursor potential in vitro was determined after culture of varying numbers (range, 20-180) of FL cells on the OP9 stromal cell line12,36 (kindly provided by Dr Kodama, Kyoto University, Yoshida, Japan), supplemented with interleukin-7 (IL-7) and c-kit ligand (KL). Half of the medium was removed and changed every 4 to 6 days. IL-2 was added at day 6 of culture. Individual colonies were analyzed by flow cytometry. NK cell differentiation was also assessed when FL cells were cultured in FTOC (see above, TCP potential). IL-7, KL, and IL-2 were obtained from the supernatants of cell lines transfected with the corresponding cDNAs and titrated as described previously.20 Erythroid and myeloid cell precursor potential. Cells were mixed with OptiMEM, 0.8% methylcellulose (Fluka, Buchs, Switzerland), and 10% fetal calf serum (FCS), supplemented with KL, IL-3, granulocyte macrophage colony-stimulating factor (4 ng/mL), and erythropoietin (4 U/mL). Hemoglobinized clusters of fewer than 100 cells were classified as erythroid colony-forming units. Large colonies of red cells (more than 300 cells) were counted as erythroid burst-forming units, whereas colonies containing at least 2 myeloid cell types and erythroid cells were classified as mixed colony-forming units. Numbers of these colonies were counted. Individual colonies were further analyzed by May-Grünwald-Giemsa staining. In vivo cell transfer Sorted cells from C57BL/6 (Ly5.1) embryos were injected in the retro-orbital sinus of 400-rad irradiated C57BL/6-Rag2/ c / (Ly5.2) mice. After 4 to 6 weeks, cells from the peripheral blood, BM, spleen, thymus, and the
small intestine (i-IELs) of the recipient mice were collected and
analyzed by flow cytometry.
Purification of NK cells and cytolytic assay Reconstituted Rag2/c/ mice were killed 6 weeks after transfer, and splenocytes were stained with monoclonal
antibodies (mAbs) specific for CD3 and NK1.1.
CD3NK1.1+ NK cells and
CD3+NK1.1 T cells were sorted and either
tested for their natural cytolytic activity or expanded for 7 days in
vitro (105 cells/mL) in complete RPMI medium (RPMI 1640 with 10% FCS, 105 M  -mercaptoethanol, 100 mg/mL
streptomycin, 100 U/mL penicillin) supplemented with 1000 U recombinant
hu-IL-2 (Peprotech). After 7 days, cells were harvested and a
standard chromium-release assay was performed. Briefly, YAC-1 (mouse
thymoma; H-2a) and P815 (mouse mastocytoma;
H-2d) target cells (106) were labeled with
37.105 Bq 51Cr (ICN Pharmaceuticals, Orsay,
France) for 45 minutes at 37°C. Cells were extensively washed, plated
at 2 × 103 cells/well, and mixed with different numbers
of effector cells. Effector cells were NK and T cells either freshly
isolated from splenocytes or activated by IL-2. The radioactivity
released into the cell-free supernatant was measured after 4 hours at
37°C, and the percentage specific lysis was calculated as follows:
100 × (experimental release  spontaneous release)/(maximum
release spontaneous release).
Reverse transcriptase-polymerase chain reaction Cells were lysed in TRIzol (GIBCO-BRL, Cergy-Pontoise, France), total RNA was isolated according to the manufacturer's protocol, and cDNA was prepared as described previously.15 Gene-specific primers used for PCR have been described previously. mb-1, RAG-1, -5,21 TCF-1,22 Pax-5,
GATA-3,7 IL-7R ,23 universal primer for
Ly49, amplifying (Ly49 C, E, F, G1-4),24 pre-TCR
(pT),15 IL-15R: 5'-CCAACATGGCCTCGCCGCAGCT-3' and
5'-TTGGGAGAGAAAGCTTCTGGCTCT-3'. The amount of cDNA in the
samples was carefully standardized by real-time PCR using hypoxanthine
phosphoribosyltransferase (HPRT) specific primers:
5'-CCAGCAAGCTTGCAACCTTAACAA-3', 5'-GACTGAAAGACTTGCTCGAG-3', and SYBR
Green PCR Master Mix (Applied Biosystems, Warrington, United Kingdom).
Samples were analyzed using a GeneAmp 5700 Sequence Detection System
(PE Applied Biosystems). The cDNA samples were amplified for 35 cycles
by the GeneAmp PCR System 9600 (Perkin Elmer, Foster City,
CA). Fifteen microliters of each PCR product was subjected to
electrophoresis through a 2% agarose gel and visualized after ethidium
bromide staining.
Immunoscope analysis for the detection of TCR rearrangements, PCR was performed on the indicated samples using a combination of 2 primers
that recognize sequences 5' of the D2.1 and 3' of the J2.7.10 The PCR products were then subjected to a
runoff reaction using a nested J2.5 fluorescent
primer.25 Runoff products were resolved on an automated
373A sequencer (Perkin-Elmer). The size and the intensity of each band
were recorded and then analyzed using Immunoscope
software.25,26
Characterization of FL cells at 15 days postcoitus based on the expression of B220 and c-kit The B220 (CD45R) marker is expressed at all stages of B-cell ontogeny. However, it is also expressed in a population of BM cells (fraction A)27,28 endowed with NK potential29 and on FL lymphoid progenitors.30 CD19 thus remains the most reliable marker for B-lineage commitment.29 The expression of the receptor tyrosine kinase c-kit correlates with precursor activity and has been used to identify HSCs, CLPs, and pro-T cells.31 FL cells were isolated from 15-days postcoitus (dpc) embryos, and erythroid precursors were depleted using the TER119 mAb. TER119 FL cells were then analyzed for
the expression of CD19, c-kit, and B220. The analysis
identified 6 distinct populations (Figure 1A). The
B220+CD19+ population (a) contains pro/pre-B
cells that develop in the FL.32,33 The CD19
cells separate into 5 other populations (b-f) according to the level of expression of c-kit and B220. All of the above
subpopulations were present at similar ratios in FL cells derived from
nude (nu/nu) embryos, showing that they are not generated in
the thymus (Figure 1A).
The majority of TCPs in the 15-dpc FL are present in
fraction e:
B220loc-kitsup+CD19 )
represents close to 70% of the TCPs in total FL cells and was thus further characterized. Figure 1D shows surface marker analysis of
sorted cells from population e after staining with the indicated antibodies. These precursors were
NK1.1CD90CD25CD44+.
In vitro progenitor activity of fraction e We further evaluated the developmental potential of the cells included in fraction e. Sorted cells were independently analyzed in 3 sensitive in vitro assays supporting T, B, and erythromyeloid differentiation from uncommitted multipotent precursors.34,35To evaluate the origin of TCPs within fraction e, we cultured sorted
cells from wild-type, nu/+, and nu/nu
C57BL/6 embryos under limiting dilution conditions in FTOCs. Cells
within fraction e generated T cells at a similar frequency
(approximately 1 in 20 cells) (Figure
2A), whereas fraction a
(B220+CD19+), used as negative control, did not
generate a detectable T-cell progeny (Figure 2A). This result indicates
that the TCPs present in population e are of prethymic origin.
To evaluate the B-cell differentiation potential, we cultured the same populations under clonal conditions with the stromal cell line S17 in the presence of IL-7 and KL. Sorted cells from fraction a gave rise to B-cell colonies at a frequency of approximately 1 in 4 cells (Figure 2A). In contrast, cells from fraction e failed to generate a B-lineage progeny. The absence of B cells in these cultures was confirmed by the absence of IgM-secreting cells after LPS stimulation (data not shown). The myeloid and erythroid potential of cells from fraction e was addressed by colony formation in methylcellulose. Unfractionated FL cells gave rise to colonies at a frequency of approximately 1 colony-forming unit per 500 plated cells (Figure 2B). Both homogeneous and mixed colonies were identified after May-Grünwald-Giemsa staining of cytospin preparations. Developing colonies included granulocytes and macrophages, and a few colonies also included erythroid cells (data not shown). In contrast, progenitors in fraction e failed to generate colonies in this assay even when up to 15 000 cells were plated (Figure 2B). An additional characterization of the lymphoid and myeloid differentiation potential was performed in a culture system using the stromal cell line OP912,36 and exogenous interleukins. These conditions permit the development of NK, B, and myeloid cells. Figure 2C shows that both total FL cells and sorted progenitors from fraction e were capable of giving rise to NK cells. Moreover, whereas unfractionated FL cells generated CD19+ B cells and myeloid cells, as indicated by the expression of Gr-1 and Mac-1, precursors in fraction e failed to generate these lineages (Figure 2C), ruling out that this fraction contained the previously identified uncommitted precursors present in FL.7,8 Evidence for a bipotent T/NK progenitor in FL by single-cell analysis The limiting dilution analysis allowed us to conclude that cells within fraction e contained T and NK precursors, either as a bipotent T/NK or as a mixture of NK and T committed progenitors. To address these possibilities, we investigated the differentiation potential of individual cells in fraction e. Cells were micromanipulated under direct microscopic inspection and allowed to colonize single irradiated fetal thymic lobes. A representative analysis of the cells generated in independent lobes is shown in Figure 3. Expression of T and NK cell markers was examined by gating on Ly5.1+ donor-derived cells. T cells expressing TCR or
TCR as well as NK+ cells expressing only NK1.1 were
present. NK1.1+TCR cells were also
generated.
Under these conditions, the T-cell readout frequency of sorted CD44+CD25+ fetal thymocytes was 1 in 5 cells.15 Using cells from fraction e, we observed thymic reconstitution in 5% (3 of 60) of the individual lobes. Although this efficiency remains low, lobes in which only T or NK cells developed were not observed. Under limiting dilution conditions, where more than one cell from fraction e was used in FTOC, independent NK or T-cell reconstitution was never observed. However, lobes containing only NK progeny were observed when cells from fraction f were seeded in FTOC (data not shown). Together these data indicate that both T and NK potentials in cells from fraction e were present in the same precursor, which we will define as C-TNKP (common T/NK cell progenitor), demonstrating for the first time the existence of such a bipotent cell in FL. TCR rearrangements have been
initiated in these progenitors. DNA was isolated from cells of fraction
e before and after FTOC, and rearrangements of the TCR genes were
examined by PCR. As shown in Figure 4A,
D-J rearrangements were not observed in cells from fraction e. In
contrast, a diverse pattern of rearrangements was observed when these
progenitors were cultured in irradiated fetal thymic lobes. These data
suggest that rearrangements in the TCR locus occur after thymic
seeding and commitment to the T lineage. Recent
findings,37 showing that the NK potential is maintained in
thymic progenitors until TCR gene rearrangements occur, support
our conclusion.
We analyzed by reverse transcriptase-polymerase chain reaction
(RT-PCR) the expression of genes that play a role in early B, T, or NK
cell differentiation. As shown in Figure 4B, sorted cells from fraction
e failed to express B-lineage-specific genes ( B220loc-kit+CD19 cells
into sublethally irradiated mice carrying mutations in both the
Rag2 and the common cytokine receptor c
genes.39 The additional absence of NK cells in the
Rag2/c/ mice40 (Figure
5A) compared with
Rag2/ mice makes them ideal hosts to analyze the
potential to generate B, T, and NK cells. Rag2/c/
mice were intravenously injected with 7 × 104
TER119 FL cells or 104 cells from fraction e,
both numbers corresponding to 500 TCPs in the respective
populations.
The analysis of blood samples from mice injected with total FL cells
revealed the presence of TCR The analysis of BM cells showed a striking difference in the B and
myeloid repopulation capacity of the 2 sets of reconstituted mice
(Figure 5C). In support of the in vitro data, fraction e was devoid of
B and myeloid cell precursor activity, as shown by the total absence of
B220+IgD+ and Gr-1+ cells, which
were generated only from unfractionated 15-dpc FL (Figure 5C). It is
interesting to note that a minor B220lo population was
detected in mice reconstituted with fraction e; those donor-derived
cells were shown to be
B220loIgD To exclude the possibility that an early transient wave of B-cell
repopulation was generated in the mice receiving
B220loc-kit+CD19 Evidence for the existence of T cells of extrathymic origin within the
intestinal epithelium41 led us to ascertain whether the
progenitors in fraction e could contain precursors for this T-cell
subset. Rag2/ Population e generates cytolytic NK cells in vivo The spleens of mice reconstituted with population e contained a subset of cells phenotypically indistinguishable from mature NK cells that we tested for their cytolytic activity. Splenic NK1.1+CD3 but not
CD3+NK1.1 cells generated in vivo from
population e exhibited low but detectable levels of cytolysis against
YAC-1 cells (Figure 6A). IL-2-activated CD3NK1.1+ cells showed strong cytolytic
activity against NK-sensitive YAC-1 thymoma cells (Figure 6A) but not
against NK-resistant P815 mastocytoma cells (Figure 6B). The cytolytic
activity of NK cells generated in vivo from population e was comparable
to that displayed by NK cells isolated from control B6 mice (Figure
6B). These data demonstrate that the
B220loc-kit+CD19
subpopulation can give rise to functional NK cells in vivo.
We have quantitatively characterized TCPs in different subsets of FL cells. This analysis allowed us to isolate a novel population of hematopoietic progenitors that represent the majority (70%) of TCPs in 15-dpc FL. Cells within this population lack in vitro and in vivo potential for B and erythromyeloid differentiation. These precursors were shown to differentiate into T and NK cells by single-cell fate analysis. Moreover, precursors uniquely committed to either the T or NK lineage were not detected in single-cell or limiting dilution assays, arguing for the absence of more restricted progenitors among this subset. This conclusion is supported by recent studies indicating that T-lineage restriction from a bipotent p-T/NK occurs in the thymus.11,37 Consistent with this notion, the cell population described here is of prethymic origin. We designate this cell population as common T/NK progenitor (C-TNKP). We show that the frequency of T-cell generation in vitro in FL-derived
C-TNKP cells is approximately 1 in 20 cells. It should be noted that a
pure population of TCPs (ie, CD44+CD25+
prothymocytes) has a plating efficiency of 1 in 5 cells in this assay.
This result indicates that not all cells in FTOC conditions colonize
the thymic lobes, and furthermore not all cells will efficiently
interact with the thymic stroma to initiate the T-cell differentiation
program. Thus, the true frequency of TCPs in the population we describe
could be much higher. It should also be pointed out that the frequency
of T-cell generation obtained here is comparable to that obtained for
CLPs from BM (1:21).5 The cloning efficiency for B-cell
precursor detection in vitro is comparable to that found for T cells in
FTOCs. Thus, 1 in 4 CD19+ FL cells can generate
B-cell colonies in the presence of stromal cells, KL, and IL-7. The
fact that we are comparing potentials of differentiation using tests
with similar plating efficiencies reinforces our conclusion that B-cell
differentiation potential is not retained in
B220loc-kit+CD19 When injected into Rag2/ The analysis of i-IELs of the reconstituted mice showed the presence of
donor-derived cells that reconstituted this particular environment with
ratios of Although sharing the common capacity to differentiate into both T and
NK cells, the population described here expressed a different set of
surface markers compared with the p-T/NK previously identified
in the fetal thymus, blood, and spleen but absent from FL, the major
hematopoietic organ during embryonic life.12 These were shown to be NK1.1+, CD117lo, and
CD90+; in contrast, the FL C-TNKPs described here are
NK1.1 Gene expression analysis revealed additional differences between these
populations. NK1.1+CD90+CD117lo
precursors expressed the T-cell-specific transcript pT
A continuous flow of immigrants seems necessary to ensure a constant T-cell generation in the thymus.46 It is conceivable that multiple cell types such as HSCs, CLPs, and p-T/NK are involved in this process. Our own previous results showed that the fetal thymus is seeded by increasing numbers of TCPs.15 The quantitative data presented here indicate the following: (1) During mid-gestation, the major population of FL cells endowed with T-cell differentiation potential are committed to the T/NK lineage; (2) this population can reconstitute both the conventional thymic and the extrathymic T-cell subsets; and (3) both by surface marker and gene expression pattern, they differ from previously described thymic and blood-derived pT/NK cells. We propose that if constant thymic immigration occurs, the population described here likely constitutes a major component of this process. The isolation of a homogeneous population of prethymic bipotent T/NK precursors will allow the identification of genes involved in early stages of T-cell commitment and differentiation and eventually will allow us to understand the molecular basis for thymic immigration. Moreover, the capacity of a human counterpart of this cell population to reconstitute efficiently the T-cell compartment could be used to prevent lymphopenia following stem cell transplantation and could therefore be of high therapeutic interest.
We thank Mathias Haury, Anne Louise for cell sorting, Pablo Pereira for advice in the i-IEL preparations, and Laurent Boucontet for help with real-time PCR analysis.
Submitted February 21, 2001; accepted June 6, 2001.
The Unité du Développement des Lymphocytes is supported by grants from the ANRS and "Ligue Nationale Contre le Cancer" as a registered laboratory. I.D. is supported by a fellowship from the Ligue Nationale Contre le Cancer.
The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked "advertisement" in accordance with 18 U.S.C. section 1734.
Reprints: Iyadh Douagi, Unité du Développement des Lymphocytes, Département d'Immunologie, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris, Cedex 15, France; e-mail: idouagi{at}pasteur.fr.
1. Akashi K, Traver D, Miyamoto T, Weissman IL. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature. 2000;404:193-197[CrossRef][Medline] [Order article via Infotrieve]. 2. Shortman K, Wu L. Early T lymphocyte progenitors. Annu Rev Immunol. 1996;14:29-47[CrossRef][Medline] [Order article via Infotrieve]. 3. Rodewald HR, Fehling HJ. Molecular and cellular events in early thymocyte development. Adv Immunol. 1998;69:1-112[Medline] [Order article via Infotrieve].
4.
Spangrude GJ, Heimfeld S, Weissman IL.
Purification and characterization of mouse hematopoietic stem cells [published erratum appears in Science. 1989;244:1030].
Science.
1988;241:58-62 5. Kondo M, Weissman IL, Akashi K. Identification of clonogenic common lymphoid progenitors in mouse bone marrow. Cell. 1997;91:661-672[CrossRef][Medline] [Order article via Infotrieve].
6.
Abramson S, Miller RG, Phillips RA.
The identification in adult bone marrow of pluripotent and restricted stem cells of the myeloid and lymphoid systems.
J Exp Med.
1977;145:1567-1579 7. Kawamoto H, Ikawa T, Ohmura K, Fujimoto S, Katsura Y. T cell progenitors emerge earlier than B cell progenitors in the murine fetal liver. Immunity. 2000;12:441-450[CrossRef][Medline] [Order article via Infotrieve].
8.
Kawamoto H, Ohmura K, Fujimoto S, Katsura Y.
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.
1999;162:2725-2731 9. Rodewald HR, Moingeon P, Lucich JL, Dosiou C, Lopez P, Reinherz EL. A population of early fetal thymocytes expressing Fc gamma RII/III contains precursors of T lymphocytes and natural killer cells. Cell. 1992;69:139-150[CrossRef][Medline] [Order article via Infotrieve]. 10. Rodewald H-R, Kretzschmar K, Takeda S, Hohl C, Dessing M. Identification of pro-thymocytes in murine fetal blood: T lineage commitment can precede thymus colonization. EMBO J. 1994;13:4229-4240[Medline] [Order article via Infotrieve].
11.
Carlyle JR, Michie AM, Furlonger C, et al.
Identification of a novel developmental stage marking lineage commitment of progenitor thymocytes.
J Exp Med.
1997;186:173-182 12. Carlyle JR, Zuniga-Pflucker JC. Requirement for the thymus in alphabeta T lymphocyte lineage commitment. Immunity. 1998;9:187-197[CrossRef][Medline] [Order article via Infotrieve]. |
Top
Abstract
Introduction
) and
B220loc-kit+CD19
) from wild-type C57BL/6 mice were cultured in methylcellulose, and
erythromyeloid colonies were counted on day 7. (C) The same populations
as in (B) were cultured on OP9 stromal cells in medium supplemented
with IL-7, KL, and IL-2 for B, myeloid, and NK cell development. Cells
were stained and analyzed by flow cytometry after 10 days of
culture.
