PeproTech Inc.
Originally published online as doi:10.1189/jlb.1107755 on February 12, 2008

Published online before print February 12, 2008
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental figures and tables
Right arrow All Versions of this Article:
jlb.1107755v1
83/5/1145    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Desanti, G. E.
Right arrow Articles by Golub, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Desanti, G. E.
Right arrow Articles by Golub, R.
(Journal of Leukocyte Biology. 2008;83:1145-1154.)
© 2008 by Society for Leukocyte Biology

Identification of CD4int progenitors in mouse fetal spleen, a source of resident lymphoid cells

Guillaume E. Desanti, Ana Cumano and Rachel Golub1

Unité du Développement des Lymphocytes, INSERM U668, Institut Pasteur, Paris Cedex, France

1Correspondence: Unité du Développement des Lymphocytes, INSERM U668, Institut Pasteur, 25, rue du Dr. Roux, 75724 Paris Cedex 15, France. E-mail: rgolub{at}pasteur.fr


arrow
ABSTRACT
 
Hematopoiesis occurs in different tissues during adult and fetal life. Splenic hematopoiesis arises in the fetal period until the first weeks of life. We have analyzed the hematopoietic progenitor content of the fetal spleen (FS) at the embryonic days 14.5–15.5. We first demonstrate that the hematopoietic content of the FS differs largely from its fetal liver (FL) counterpart. The difference mainly concerns the distribution of the different pool of progenitors, as most of the splenic progenitors are comprised in the lineage–Sca1–cKitlo contrary to the FL. We have divided the fetal hematopoietic pool into smaller fractions to enable characterization of the earliest lymphoid progenitors. Among the lymphoid progenitors that already represent a rare population, we were able to separate a population, respectively, enriched in B or T/NK progenitors. Lineage restriction of the different developmental intermediates was tested by clonal assays. We propose a model for fetal splenic hematopoietic progenitors and their distribution.

Key Words: hematopoiesis • cell differentiation • fetal development.


arrow
INTRODUCTION
 
During fetal life, the liver is colonized at embryonic day 10.5 (E10.5) by hematopoeitic stem cells (HSC) generated in the para-aortic splanchnopleura/aorta-gonad-mesonephros (AGM) region [1 ]. From E12 to birth, the fetal liver (FL) is the major site of embryonic blood production. Fetal and adult hematopoiesis occurs from HSC to mature cells through a hierarchy of defined, lineage-restricted progenitors that are isolatable on the basis of their cell surface phenotype. HSC are comprised within the lineage (Lin)–Sca1hicKithi population, commonly named LSK [2 ]. In the adult bone marrow (BM), the LSK have been further subdivided into the long-term HSC, the short-term HSC, the multipotent progenitors (MPP), the lymphoid-primed MPP, and the early lymphoid progenitors (ELP) [3 4 5 6 ]. These different LSK subsets are discriminated by their differential expression of fetal liver tyrosine kinase 3 (Flt3), CD34, {alpha}2 integrin, VCAM-1, and RAG molecules [7 , 8 ]. Although each of these LSK subsets retains the capacity for myeloid and lymphoid differentiation, they differ in their self-renewal potential. The long-term HSC population self-renews for the lifetime of the mice, whereas the short-term HSC, MPP, and ELP show a transitory or no self-renewal potential. The MPP would give rise to the identified common myeloid progenitor (CMP) and common lymphoid progenitor (CLP), which differentiate exclusively to specific cell lineages [9 , 10 ]. Studies about FL hematopoietic progenitors have isolated the phenotypic counterparts of adult CMP and CLP as distinct populations at E12.5–14.5 [11 , 12 ]. The fetal equivalents of adult CMP and CLP hold differences in their potential. The FcRloCD34+ CMP generates the Fc{gamma}RhiCD34+ GM lineage-restricted progenitor (GMP) and Fc{gamma}RloCD34– megakaryocyte/erythrocyte lineage-restricted progenitor (MEP) and shows a weak but significant B cell potential [12 ]. Despite a common IL-7R{alpha}+cKitloSca1lo phenotype with the adult CLP, this fetal progenitor still retains myeloid potential [10 , 11 ]. Studies have identified and compared the lymphoid progenitors in murine FL and adult BM thanks to the use of transgenic RAG1GFP mice [5 , 13 ]. The purification of BM ELP as cKithi cells has revealed that cKit progressively declines with normal lineage progression [14 ]. The number of cKit+ cells in the FL increases up until E15 and progressively decreases thereafter. The diminution of cKit is a reciprocal function of GFP density for most putative lymphoid progenitors in embryos. The expression of Sca1 on fetal progenitors evolves similarly to cKit. More restricted progenitors have also been identified from the FL, such as a bipotent myeloid/B and T/NK progenitor [15 , 16 ].

The CD4 cell surface molecule is expressed by differentiated cells, such as T cells and a particular subset of dendritic cells. CD4 is also expressed by hematopoietic progenitors found in the FL, the mesenteric lymph node anlagen, the adult BM, blood, and thymus [5 , 17 18 19 ].

The fetal spleen (FS) is colonized by HSC and hematopoietic progenitors from E13.5 and is considered as a transient hematopoietic organ, as it becomes an adult immune organ [18 , 20 21 22 23 ]. The adult spleen is well known for its secondary immune organ function, its capacities for removing senescent erythrocytes, and for filtering the blood (for review, ref. [24 ]). However, little is known about its properties during the fetal life. The FS can be isolated from the pancreas as an individual splenic rudiment at E11.5 [25 ]. The early developmental processes are independent of its hematopoietic colonization. Most of the hematopoietic progenitors that reside in the FS as well as their potential are unidentified. The only precursors characterized are the HSC and a common T/NK progenitor [26 , 27 ]. We have recently demonstrated that the FS restricts HSC commitment toward the myeloid lineage, whereas lymphoid differentiation is sustained by committed progenitors [23 ].

In this study, we sought to determine whether the hematopoietic content and kinetic of the FS are similar to the FL. Therefore, we searched for counterparts of FL progenitors by screening important markers. We found that the repartition of cKit and Sca1 markers was different. We show that the immature compartment of the FS has a reduced frequency of LSK and Lin–Sca1–cKithi, CMP, GMP, and MEP as compared with the FL.

We report that despite this dissimilarity between FL and FS progenitors, it was possible to isolate lymphoid progenitor populations with significant phenotypic similarities. We show through the use of flow cytometry, clonal, and in vivo assays for progenitor potentials that FS is enriched for a CD4intcKitloLin– population with high lymphoid-differentiation potential. After purification, we analyzed the phenotype and studied the hematopoietic differentiation potential of this population. We show that it is able to give rise to B, T, NK, and less efficiently, myeloid cells in vitro. Clonal in vitro assays showed that the FS CD4int population is composed of different progenitors mostly committed to lymphoid cell fates. These CD4int progenitors were further subdivided by their differential expression of a RAG2GFP transgene, and the different hematopoietic potentials were separated according to different GFP levels. The acquisition of RAG2 by the progenitors is concomitant to that of IL-7R{alpha} expression, the loss of myeloid potential, and restriction to the B lineage.


arrow
MATERIALS AND METHODS
 
Mice
C57BL/6-CD45.1 mice (CDTA, France), C57BL/6-CD45.2 mice (Charles River France, Lyon), and C57BL/6-Rag2/{gamma}c–/– CD45.1 mice [28 ] were bred in our animal facility (Institut Pasteur, Paris, France). Friend virus B-type (FVB) RAG2GFP male mice [29 ] were mated with C57BL/6 female mice to generate RAG2GFP embryos. Pregnant female Swiss nude/+ mice were provided by Charles River France, and nude/nude embryos were segregated by their absence of thymus. The vaginal plug observation was considered as the E0.5. All animal experiments were approved by the Institutional Animal Care and Use committees in accordance with the applicable national regulations.

Cell preparation and FACS staining
Fetal tissues were dissociated with a 26-gauge 3/8 needle. FLs were erythrocyte-depleted by 40%/70% Percoll gradient (2600 rpm, 30 min) or by streptavidin-magnetic beads depletion (Dynabeads M-280, Dynal, Norway) after Ter-119-biotin staining (Becton Dickinson-PharMingen, San Diego, CA, USA). Cells were counted after trypan blue exclusion of dead cells. Flow cytometry analyses were performed in an upgraded LSR (Becton Dickinson, San Jose, CA, USA) with the CellQuest software (Becton Dickinson) or the FlowJo software (Tree Star, Ashland, OR, USA). The following antibodies, all from PharMingen [biotinylated or coupled with different FITC, PE, allophycocyanin (APC), or PE-Cy7], were used: CD45.2 (clone 104), CD11b/Mac1 (M1/70), Gr1 (RB6-8C5), CD11c (HL3), CD45R/B220 (RA3-6B2), CD19 (1D3), Ter-119, CD34 (RAM34), CD31 (MEC13.3), NK1.1 (PK136), TcRβ (H57-597), {gamma}{delta} TcR (GL3), Sca1 (E13-161.7), CD117/cKit (2B8), CD3{epsilon} (145-2C11), CD4 (L3T4), CD122 (TM-β1), {alpha}4β7 (DATK32). The antibodies CD127/IL-7R{alpha} (A7R34) and CD45.1 (A20) were purchased from eBioscience (San Diego, CA, USA). The lineage mix consists of Gr1, Ter119, B220, CD19, CD11c, NK1.1, and CD3{epsilon} antibodies coupled to PE. Dead cells were eliminated by propidium iodide (PI) exclusion.

Cell sorting
Cells were separated using a MoFlo sorter (DakoCytomation, Denmark). Cells were recovered in Eppendorf tubes or directly into wells of culture plates using the Cyclone computer program. Dead cells were eliminated by PI exclusion. Ter119-depleted FL cells were stained with lineage PE antibody mix, Sca1-FITC, and cKit-APC to sort LSK cells. FS cells were stained with the lineage-PE antibody mix, CD4-FITC, or PE-Cy7 and Mac1-APC to sort CD4int Mac1–/loLin– cells.

Cell culture conditions
All experiments were done in 96-well plates at 37°C, 5% of CO2, and in a culture medium consisting of OptiMEM, 10% FCS, 100 U/ml penicillin, 100 µg/ml streptomycin, 5 x 10–7 M 2-β-ME (Gibco, Grand Island, NY, USA).

Clonal assays
One cell was directly plated using the MoFlo sorter, and culture conditions were identical to those used for the limiting dilution assay. In this study, OP9 or OP9 expressing the Notch ligand {Delta}-like1 (OP9-Dl1) stromal cells was seeded in 96-well plates (5000 cells/well). The culture medium was supplemented with saturating amounts of cKit ligand (cKitL), Flt3 ligand (Flt3L), IL-2, IL-3, GM-CSF, and IL-7. OP9 was used for the analysis of B/NK/myeloid potential. Scores were assigned to B cell (CD19+B220+ or CD19+RAG2–/+), NK (NK1.1+CD122+ or NK1.1+RAG2–), or myeloid colonies (CD11b+) on days 14–18 by flow cytometry. On OP9-Dl1, T cell (CD3{epsilon}+TcR+) and NK (NK1.1+CD3{epsilon}–) cell colonies were assayed identically. For each experiment, a minimum of 180 wells was analyzed.

FS and FL organ culture
E15.5 FS and FL were explanted, placed on a 0.8-µm filter (ATTP, Millipore, Bedford, MA, USA) for 4 days on complete medium. Then, cells suspensions were obtained from FS and FL organ cultures, counted, and analyzed by flow cytometry in parallel to ex vivo E15.5 FS and newborn spleen (NBS).

In vivo reconstitution
Between 2000- and 2500-sorted FS CD4intLin– (CD45.2), cells were injected into 600-rad, irradiated 3- to 7-week-old Rag2/{gamma}c–/– (CD45.1) mice. After 7–8 weeks, cells from the spleen, BM, and liver were harvested and analyzed by flow cytometry for donor origin. Livers were previously perfused with PIRM (Gibco), 2% FCS, to remove cells from the peripheral blood. Lymphocytes were recovered after centrifugation in 35% Percoll. Erythrocyte elimination was done by incubation with ammonium chloride solution. For each experiment, littermate-noninjected Rag2/{gamma}c–/– mice were used as a negative control. Moreover, Rag2/{gamma}c–/– mice injected with 2000 FL LSK cells or with one-third of FL embryonic equivalent were used as positive controls.


arrow
RESULTS
 
Kinetic of FS hematopoiesis
To determine the hematopoietic content of the FS, we first determined the proportion of splenic erythrocytes and leukocytes at different developmental time-points (Table 1 ). The splenocyte percentages were obtained by flow cytometry using the combination of anti-Ter119 and anti-CD45 antibodies to, respectively, detect erythrocytes and leukocytes at E13.5, E14.5, and E15.5. At these developmental stages, most of the splenic erythroid cells are CD71+ precursors [30 ]. Hematopoietic cells represent a minor population of splenocytes (21–33%) at E13.5 and increase to be in the majority (at least 53%) as soon as E14.5. At E15.5, they are the essential lineage present (Table 1) . Leukocytes become predominant at E15.5 stage. Moreover, this transition in the CD45+ cell representation is accompanied by a threefold increase of the splenocyte absolute number (Table 1) .


View this table:
[in this window]
[in a new window]

  		Table 1. Percentages and Numbers of Splenocytes According to the Mouse Developmental Stage

To further determine the kinetic of splenic hematopoiesis, we observed the ratio of hematopietic progenitors (Lin– cells) during fetal development. As FS at E14.5 are mainly composed of progenitors, and inversely, at E16.5, the key transitional stage where most of the differentiation process should occur, is E15.5 (Fig. 1A ). Hence, we defined this stage as central to analyze the splenic hematopoiesis, considering it as the best compromise in terms of number of cells available per embryo and proportion of progenitors per organ.


Figure 1
View larger version (48K):
[in this window]
[in a new window]

  		Figure 1. The hematopoietic content of FS. (A) Leukocytes (CD45+) from E14.5, 15.5, and 16.5 FS were analyzed for the expression of markers expressed by lineage-committed cells (Lin: Ter-119, Gr1, B220, CD11c, CD19, NK1.1, CD3{epsilon}). (B) Sca1 and cKit expression from E15.5 FS and FL Lin–PI– gated populations. The livers were previously depleted with Ter119-biotin antibody and streptavidin-magnetic beads. Numbers indicate the percentage of cells in the gates.

FS and FL Lin– cells were scored for the expression of Sca1 and cKit (Fig. 1B) . In the FL, 1.6% of Lin– cells belong to the LSK population, whereas the majority of them is cKithiSca1– (Fig. 1B) . In contrast, no LSK cells are detectable in the FS, where the Lin– population mostly consists of cKitlo and negative cells (82.8%; Fig. 1B ). It seems that E15.5 FS are mainly composed of progenitors that have down-regulated cKit at the surface. The difference of cKit expression in the Lin– hematopoietic progenitors between FL and FS directed us to look for the lymphoid progenitors and their hierarchy in FS.

Identification of the FS hematopoietic progenitors
To isolate the FS hematopoietic progenitors, the E15.5 FS CD45+Lin– population was screened for the expression of CD34 (Fig. 2A ). As the major source of hematopoietic progenitors during fetal life, the FL was analyzed in parallel. FS and FL progenitors are separated into CD34+ and CD34–. Contrary to FL, most of FS Lin+ cells are CD34lo. The CD4 marker identifies a population of progenitors that expresses CD34 differentially in the FS and FL. Whereas the Lin–CD4int population is CD34lo in the FS, it is clearly CD34+ in the FL (Fig. 2A and Supplemental Fig. 1). A population of CD4hiCD34–Lin– cells is only present in the spleen and probably represents the lymphoid tissue inducer cells that are necessary to generate the secondary lymphoid organs [11 ]. In these two organs, CD4int cells express CD31 (Fig. 2B) . Most of the FL Lin–CD34+CD4int cells in the FL are cKithi. Inversely, the low expression of CD34 in the FS is associated to the absence of cKithi cells (Fig. 2C) . This Lin–CD4intCD34lo population is representative of the overall bias toward a majority of cKit-/lo progenitors in the FS (Fig. 1) . At E15.5, the CD4intLin– population represents 3.5% of the FS hematopoietic cells and 5.3% of those from the FL, and it is also found in the blood (Supplemental Fig. 2). These populations are already present in the FS and the FL of E14.5 embryos (data not shown).


Figure 2
View larger version (47K):
[in this window]
[in a new window]

  		Figure 2. The FS and FL progenitor populations are composed of CD4int cells. (A) E15.5 FS and FL were analyzed for expression of the CD34 and lineage (Lin: Ter-119, Gr1, B220, CD11c, CD19, NK1.1, CD3{epsilon}) surface markers. CD4 and CD34 expression is analyzed from lineage-negative cells. The livers were previously erythrocyte-depleted with Ter119-biotin antibody and streptavidin-magnetic beads. (B) E15.5 FS and FL were analyzed for expression of the CD4 and CD31 surface markers. The livers were previously erythrocyte-depleted by 40%/70% Percoll gradient. (C) CD4intLin– were gated and studied for the expression of Sca1 and cKit surface markers. Numbers indicate the percentage of cells in the gates.

RAG2 expression subdivides FS CD4int cells into subsets with distinct differentiation potential
RAGGFP mice have been used to further characterize the hematopoietic progenitors. The RAG1 and RAG2 proteins are expressed at the ELP and CLP stages [31 ]. In a RAG1GFP mouse, it has been shown that the progenitors restricted toward the lymphoid lineage express RAG1 and the IL-7R{alpha} (CD127) proteins concomitantly [5 , 13 ]. Moreover, the RAG1 expression level also discriminates T from B progenitors that are contained in the cKitloSca1– population from the FL [13 ]. In a RAG2GFP mouse, the ELP have been shown to express low levels of RAG2 [31 ]. We used RAG2GFP transgenic mice [29 ] to discriminate the FS precursors contained in the CD4intLin– population. FS and FL CD4intLin– cells from RAG2GFP embryos could be subdivided into three subsets depending on their level of GFP expression (Fig. 3A ). Likewise, these three subsets of the CD4intLin– cells are found in the blood (Supplemental Fig. 2). The levels of IL-7R{alpha} and RAG2 increase concomitantly: RAG2– cells are mainly IL-7R{alpha}–, RAG2lo cells are IL-7R{alpha}–/lo, and RAG2hi cells are IL-7R{alpha}hi (Fig. 3A) . It should be noted that a small fraction of FS CD4intLin–RAG2– cells expresses the IL-7R{alpha}.


Figure 3
View larger version (35K):
[in this window]
[in a new window]

  		Figure 3. The FS CD4int population is composed of several hematopoietic progenitors. (A) FS and FL CD4+Lin– cells were analyzed by flow cytometry from E15.5 RAG2GFP embryos. Numbers indicate the percentage of cells in the gate. For each RAG2 subset, the IL-7R{alpha} expression is shown (histograms). (B) The hematopoietic potential of RAG2GFP FS CD4intLin– cells was tested after culture on a OP9 or OP9-Dl1 stromal cell layer with a cytokine mix (cKitL, Flt3L, IL-2, IL-3, IL-7, GM-CSF). Macrophage (Mac-1hi), B (CD19+RAG2–/+), and NK (NK1.1+RAG2–) cell differentiation has been assayed after 2 weeks of culture on OP9 (left dot plots). NK (NK1.1+CD3{epsilon}–RAG2–) and T (TcR{alpha}β/{gamma}{delta}+CD3{epsilon}+RAG2-/+) cell differentiation assays were performed after 2 weeks of culture on OP9-Dl1 (right dot plots). (C) Clonal assays were performed for each CD4intLin– RAG2 subset. The macrophages (Mac), B, NK, and T cell differentiation were checked by FACS. The histograms are the percentage of wells where a progeny is growing, with the defined lineages, when we hypothesize that a cell has seeded each well during the MoFlow cell sorting. On OP9, a progeny is obtained in: 31.5 (±9.7)% of wells seeded with the RAG2– subset, 58.6 (±12.6)% of wells seeded with the RAG2lo subset, 54.6 (±11.2)% of wells seeded with the RAG2hi subset. On OP9-Dl1, a progeny is obtained in: 21.4 (±3.3)% of wells seeded with the RAG2– subset, 43.1 (±8.9)% of wells seeded with the RAG2lo subset, 12.7 (±7.6)% of wells seeded with the RAG2hi subset. Histograms are the mean of three independent experiments. (D) Clonal assays of CD4intLin– cells sorted from C57BL/6 embryos were performed on OP9 during 3 days, followed by a redistribution of 50% of each well content on OP9 and OP9-Dl1 conditions. The B, NK, and T cell potential was assessed 12 days later by flow cytometry. The histograms are the relative representation of each progeny obtained [relative representation=(percentage of wells with defined lineages/percentage of wells where a progeny is growing) x 100].

The RAG2 subsets were studied for their hematopoietic potential by clonal assays on OP9 and OP9-Dl1 (Fig. 3B) . On OP9, the B, NK, and myeloid potential are, respectively, evaluated as the expression of CD19, NK1.1, and Mac1 by cytometry. On OP9-Dl1, NK and T cells were detected using anti NK1.1, CD3{epsilon}, and TcRβ and TcR{gamma}/{delta} antibodies (Fig. 3B) .

The RAG2– population exhibits a mix of pluri- and mono-potent progenitors. The RAG2lo population is mainly composed of progenitors engaged in the T and NK cell pathways. Myeloid potential is still observed in a small fraction of RAG2lo cells. The RAG2hi subset is mainly composed of B cell-restricted progenitors. When cells from the RAG2hi subset were clonally cultured on OP9-Dl1, few of them were able to grow (e.g., 12.7%±7.6). However, as OP9-Dl1 is a driving stroma for T cell development, more than 90% of the cells that have grown are T/NK bipotent. These cells probably represent the 5% of NK progenitors estimated on OP9 (Fig. 3C and Supplemental Fig. 3). Hence, RAG expression separates T/NK from B progenitors in FS.

To assess the capacity of the T/NK progenitors to give rise to B cells, total Lin–CD4int cells were cultured by a two-step clonal assay. Cultures were initiated on OP9 cells during the first 3 days, and then the progeny was divided on OP9 and OP9-Dl1 cells for 2 weeks (Fig. 3D) . These experiments point to the existence of a B/T/NK potential in the CD4intLin– population. It also confirms the main presence of unipotent, B-committed cell progenitors (CD4intRAG2hi).

The FS CD4intLin– population is present in the same proportion in athymic nu/nu mice. Limiting dilution assays led to the same ratio of B, T, and NK progenitors from nu/nu mice compared with nu/+ and wild-type mice (data not shown). Thus, the FS CD4intLin– population does not contain progenitors of thymic origin.

B cell progenitors were observed as RAG2hiLin– cells at E14.5-like FS B (CD19+B220+) lymphocytes that can be isolated as RAG2hiLin+ cells (data not shown). NK cells with an immature NK1.1+CD122+Mac-1– phenotype are present in E15.5 FS but not before (data not shown).

In conclusion, RAG2 expression allows the discrimination of three distinct subsets of the FS CD4loLin– progenitors. The RAG2– subset mainly contains multilineage progenitors and unipotent myeloid progenitors. The RAG2lo subset is enriched in NK/T precursors, and the RAG2hi subset represents a B cell-committed population. These results are consistent with those discriminating FL progenitor potential in the RAG1GFP mice [13 ].

FS CD4int cells reconstitute the B and NK compartments of Rag2/{gamma}c–/– mice
In vivo reconstitution assays are commonly used to confirm the hematopoietic potential of progenitors previously studied in vitro. Sorted FS CD4intLin– cells were injected into sublethally irradiated Rag2/{gamma}c–/– mice (Fig. 4A ). We used Rag2/{gamma}c–/– injected with sorted FL LSK or total FL cells as reconstitution positive controls. At 7–8 weeks postinjection, the BM reconstitution is representative of a transient engrafment by committed progenitors. A small reconstitution of the myeloid (Mac1hiNK1.1–; Mac1hiGr1hi) compartment was obtained in the BM and spleen (Fig. 4B) . The B and NK cell reconstitution observed in the BM uncovers an absence of hematopoietic progenitors, respectively, B220+CD19– or NK1.1+Mac1– cells.


Figure 4
View larger version (59K):
[in this window]
[in a new window]

  		Figure 4. FS CD4intLin– cells reconstitute the lymphoid and the myeloid compartments of Rag2/{gamma}c–/– mice. (A) CD4intLin– cells are isolated from B6-CD45.2 embryos and injected into Rag2/{gamma}c–/– CD45.1+-irradiated mice. The numbers between brackets indicate the variation of gated cell percentage found in five independent experiments. (B) After 7–8 weeks, spleen, BM, and liver were analyzed by flow cytometry. Donor-derived cells (CD45.2+) are gated and their hematopoietic phenotype analyzed. Numbers indicate the percentage of cells in the gates. As a positive control, Rag2/{gamma}c–/– mice were injected with one-third of FL embryonic equivalent or 2000 FL LSK cells, and results were compared with mice injected with 2000–2500 FS CD4intLin– cells (FS CD4int). The chimerism observed is: 0.10% (±0.13) CD45.2+ cells in the BM, 1.12% (±1.68) CD45.2+ cells in the spleen for injection with FS CD4int cells, 47.8% (±22.5) CD45.2+ cells in the BM, and 70.4% (±8.9) CD45.2+ cells in the spleen for injection of HSC FL-derived (three) mice. The data presented here are representative of three independent cell-sorting and injection experiments.

Donor-derived B (CD19+B220+) cells were found in the spleen and BM of approximately 60% of recipient mice (Supplemental Tables 1 and 2). Donor-derived NK cells (NK1.1+CD3{epsilon}–) are detected in the spleen and liver of all the recipients (Supplemental Tables 1 and 2). The FS CD4int population is apparently biased in its capacity to reconstitute the NK compartment with a high percentage of NK cells in the BM and liver (more than 60%, in average). T cells (CD3{epsilon}+NK1.1–) could hardly be found in the spleen and liver. The low percent of donor T cells in all organs indicates a poor potential toward this lineage in vivo.

In conclusion, FS CD4intLin– sorted cells are efficient in reconstituting transient, mature B and NK compartments. In each organ studied, the chimerism observed is weak when compared with a FL population enriched in HSC. Indeed, the percentages and absolute numbers of FS donor cells are low when compared with the HSC FL-derived cells (Supplemental Tables 1 and 2). The FS CD4intLin– population, therefore, represents a combination of early progenitors already engaged. The absence of myeloid engraftment by precursors supports the probable absence of HSC from the FS CD4int cells.

In the absence of blood influx, the FS environment sustains the lymphoid and the myeloid cell differentiation
To assess the in situ differentiation of splenic Lin– progenitors, we performed FS explant organ culture (FSOC). The E15.5 fetal organs were cultured in toto and developed in the absence of blood influx for 4 days. In the FSOC, the percentage of Lin– cells decreases by more than twofold compared with E15.5 FS (Fig. 5A ) simultaneously to an increase of mature cells, suggesting that progenitors differentiate in the explants.


Figure 5
View larger version (37K):
[in this window]
[in a new window]

  		Figure 5. FS CD4int cell progenitors do not maintain in situ. (A) Evolution of the Lin– progenitor (bars) population in E15.5 FS, E15.5 FS explants after 4 days of organ culture (OC15.5+4), or from NBS (NB), analyzed by flow cytometry. The Lin– progenitors represent 44.1% (±1.9) of E15.5 FS (n=6 samples of five to six spleens), 18.9% (±1.2) of OC 15.5 + 4 (n=4 samples of six to seven FSOC), and 15.6% (±0.7) of the NB (n=4 samples of three spleens). (B) Coevolution of the CD4 and CD34 markers in the Lin– cells from the population analyzed above. Numbers indicate the percentage of the gated CD4intCD34+Lin– progenitors. (C) Absolute numbers of total hematopoietic and CD4intCD34+Lin– cells in E15.5 spleen (n=6 samples of five to six spleens), E15.5 FS explants after 4 days of organ culture (FSOC, n=4 samples of six to seven FSOC),or from NBS (n=4 samples of three spleens). These data are representative of the three independent experiments. (D) Evolution of the NK (NK1.1+) and myeloid (Mac-1+) populations in E14.5 FS, E14.5 FS explants after 4 days of organ culture (OC 14.5+4), or from E18.5 FS, analyzed by flow cytometry. Numbers indicate the percentage of cell in the gates.

The CD4intLin– population development was analyzed by cytometry (Fig. 5B) . In the FL organ cultures (FLOC), the percentage of CD4intLin– cells is stable, whereas it drops in FSOC compared with ex vivo FS (Fig. 5B) . Similarly, the absolute number of CD4int cells is also decreasing in FSOC, whereas the total number of cells is identical (Fig. 5C) . In FSOC, a decrease of CD4intLin– cells is clear, and this reduction could be a result of their differentiation in association and a poor capacity to be maintained (Fig. 5) . In contrast, the CD4intLin– population is maintained in FL explants.

We have defined the hematopoietic potential of CD4intLin– cells and underlined the capacity of the FS environment to sustain the hematopoiesis. These progenitors could differentiate into B lymphocytes [22 ] as well as NK and myeloid cells (Fig. 5D) . In consequence, the FS is composed of progenitors with B, NK, and myeloid potential, and the FS environment can sustain their differentiation.


arrow
DISCUSSION
 
We have shown that the FS is seeded early with erythrocytes and leukocytes. The FS is a hematopoietic organ as soon as E13.5. The ratio between hematopoietic progenitors and mature cells evolves during embryogenesis to reach an equilibrium at E15.5, suggesting an active differentiation process at this developmental point. We selected E15.5 as the best developmental stage to study the FS hematopoiesis, where FS progenitors are enough and not too diluted among mature cells. We have estimated that 64% of splenocytes are leukocytes at E15.5. We have shown that E15.5 FS, isolated from the circulation and cultured for 4 days, present more committed cells and less progenitors than an ex vivo E15.5 FS supporting an in situ differentiation.

The FL is considered as the most important fetal hematopoietic site, and hematopoietic progenitors have been analyzed extensively [14 ]. We examined the hematopoietic content of FS and compared it with the FL. The FS and FL harbor differences in the phenotype and proportion of the Lin– subset. The screening of cKit and Sca1 expression in FS progenitors reveals a peculiar situation. In FL, the Lin– hematopoietic progenitors are subdivided into cKithigh cells (CMP, GMP, and MEP) and LSK that represent 59% and 1.6%. However, the largest Lin– population in the E15.5 FS is cKit–/lo (83%), and the cKithigh cells and LSK, respectively, represent 12% and 0.1%. It is known that cKit is down-regulated with differentiation in the BM [32 ]. Besides, cKit levels in the FL are also declining proportionnally to the RAG1 and lineage marker acquisition [13 ]. Hence, this difference of cKit expression prompted us to characterize the FS progenitors. Lin–CD34+ cells that express intermediate levels of CD4 were isolated. These progenitors are mainly comprised in the Sca1–cKitlo fraction. This phenotype distinguishes them from HSC (LSK) and myeloid-restricted progenitors (Lin–Sca1–cKithi). FL and FS CD4intCD34+ progenitors are obviously different for the expression of cKit. The FL counterpart is reminiscent of CMP and HSC. In the FS, the same progenitors express low levels of cKit that reveals an absence of cells expressing the phenotype of well-defined precursor subsets (CMP, MEP, GMP) [12 ]. We concluded to determine the differentiation potential of FS CD4intLin– cells.

HSC have been functionally identified as a rare LSK cell subset in the FS [21 , 27 ]. Our in vivo and in vitro assays have confirmed the absence of HSC from FS Lin–CD4intCD34+. No BM engraftment was detected after injection into alymphoid recipients. Hardly any myeloid cells could be recovered after 8 weeks, and no lymphoid precursor subsets (B220+CD19– or NK1.1+Mac1–) were detected in the BM. All donor-derived lymphoid populations consisted of mature cells, showing that FS CD4intLin– cells gave rise to only one wave of differentiation. In addition, in lymphopenic Rag2/{gamma}c–/– mice reconstituted with 2000 FS CD4int cells, the myeloid chimerism was much lower than reconstitution with FL LSK. Additionally, we have shown from clonal assays that only a small percentage of CD4int cells is able to differentiate into myeloid cells. FS CD4int cells are therefore already committed, and most of them are engaged in the lymphoid lineage.

Clonal assays on OP9 and OP9Dl1 cell lines revealed that the FS CD4int population is composed of a heterogeneous population of precursors. We therefore used RAG2GFP transgenic mice [29 ] to further subdivide this population into precursors with different hematopoietic differentiation capacities. We found that RAG2 is differentially expressed in the FS CD4int population and could be separated into RAG2–, RAG2lo, and RAG2hi subsets. The clonal analyses of these subsets allowed for further discrimination of the hematopoietic progenitors. The unipotent B cell potential is largely restricted to the RAG2hi compartment, and the RAG2lo represents a majority of the T/NK precursors. The RAG2– subset is composed of pluripotent precursors, including those with the myeloid capacities, and unipotent myeloid and NK progenitors (Fig. 6 ). The FS RAG2– subset is mainly IL-7R{alpha}–, the RAG2lo subset is IL-7R{alpha}–/lo, and the RAG2hi subset is IL7R{alpha}hi. The loss of myeloid potential concomitantly with the acquisition of RAG2 and IL-7R{alpha} was also observed from FL progenitors isolated from RAG1GFP mice [13 ]. Such similarities are explained by the coordinated expression of RAG1 and RAG2 proteins [33 ]. As the FL is the main source of hematopoietic cells at E14, we propose that the FS CD4int progenitors originate from the FL and then differentiate herein. Consistent with this idea, we detected CD4intLin– cells in the blood, and we observed that part of the FL and FS CD4intLin– shares a RAG expression pattern and some surface markers such as IL-7R{alpha}.


Figure 6
View larger version (17K):
[in this window]
[in a new window]

  		Figure 6. Model for the FS CD4intLin– progenitor differentiation. The FS CD4intLin– population is characterized at E15.5 and subdivided into three subsets based on their RAG2 expression levels. The myeloid potential of the FS CD4intLin– population decreased, and the RAG2 expression increases (triangle). The RAG2– fraction is mainly composed of IL7R{alpha}– cells, despite a small population of IL-7R{alpha}hi cells. This fraction is a mix of different hematopoietic progenitors. The RAG2lo fraction is IL-7R{alpha}–/lo and mainly composed of bipotent progenitor (p)-T/NK precursors. In this fraction, we have also determined a macrophage-restricted progenitor. The RAG2– and RAG2lo subsets probably participate to the elaboration of the FS pool of macrophages and NK cells. The RAG2hi fraction is IL-7R{alpha}hi cells and essentially consists of p-B cells that likely generate the FS B cell pool. Size of circles is proportional to the progenitor frequency of each CD4intRAG2 compartment. The arrows represent the differentiation toward mature hematopoietic cells.

CD4int lymphoid progenitors cannot derive from FS HSC, as we have previously shown that FS stroma restricts the HSC fate to the myeloid lineage. Hence, irradiated FSOC that were reconstituted with AGM or FL HSC were unable to generate CD4int cells, and all donor-derived cells were F4/80+ macrophages [23 ]. Differentiation of B cells in FS is driven from already committed progenitors and probably from the FS Lin–Sca1–cKitloCD4intCD34+RAG2hi progenitors. From early studies that have not assessed the CD4 staining, it was found that the Sca1–cKitloRAG1hi FL subset exclusively generates CD19+ B lymphocytes [13 ]. We suggest that the FS RAG2hi subset is derived from its FL equivalent via the circulation. In accordance with these data, we have shown here that CD4intLin– cells are not able to maintain in FS explants, whereas they can in FL explants. It suggests that CD4intLin– cells differentiate in its FS and are not renewed by other hematopoietic progenitors present in this organ. By contrast, the CD4intLin– population present in FL explants could be maintained by the differentiation of hematopoietic progenitors coming from the HSC pool.

Among the CD4intLin– progenitors, the NK precursors are numerous, as they, respectively, represent 44% and 25% of the RAG2lo and RAG2– subsets. NK cells also represent the predominant, reconstituted compartment after injection into alymphoid mice. Moreover, most of the T cell precursors isolated among the FS CD4intLin– cells are bipotent with a NK potential. These precursors are thymic-independent, as they could be found in the same proportion, and keep the same frequency when isolated from FS of nude mice. We thus identified a new thymic-independent p-T/NK subset. The differentiation pathway of FS T/NK progenitors is classical, as the CD4+CD8+ double-positive stage is observed, and the T cells obtained could express {alpha}β or {gamma}{delta} TcRs after in vitro cultures. However, upon in vivo reconstitution, donor T cells are rare, and NK cells are found in all reconstituted mice. We propose that NK cell reconstitution is favored to T cell reconstitution as a result of a better accessibility of progenitors to the BM than to the involuting thymus of alymphoid-recipient mice (for review, ref. [34 ]).

In conclusion, we identified FS differentiation pathways, restricted to the embryonic development that is no longer accessible in adult spleen. The FS is able to attract and host lymphoid progenitors that kept myeloid potential. In this organ, we noticed that cells with the HSC phenotype are rare and that few cells corresponding to CMP, GMP, and MEP are present when compared with those from FL. It appears that only few cKithi cells are present in FS, suggesting a different dependence on cKitL or a different regulation of cKit on the cell surface.

We characterized the FS CD4int cells as ELP. FS CD4intLin– cells also express the endothelial CD31 marker. Most of the lymphoid progenitors in embryos express endothelial markers [13 ].

The CD4intLin– progenitors are not maintained in the FSOC, indicating that the maintenance and expansion of hematopoietic progenitors are not favored in this environment. It reinforces the notion that FS in contrast to FL provides a unique hematopoietic environment capable of supporting the differentiation of myeloid and lymphoid cells but is unable to sustain the expansion and survival of HSC [23 ].

The spleen is colonized early by lymphoid progenitors. The first FS lymphocytes are detected at approximately E16 and expand in situ [35 ]. We postulate that this CD4intLin– population gives rise to the majority of mature lymphoid cells found at birth in this organ. Our studies pave the way to the identification of factors important for the hematopoietic development in this environment. We will also define if progenitors in this special environment can be driven to a specific phenotype and/or function. In this context, it is particularly important to understand the implication of the fetal splenic stroma in the ontogeny of particular subsets of B lymphocytes present in the adult spleen, such as marginal zone B cells (for review, ref. [36 ]), and in the survival of the B1a subset [37 ].


arrow
ACKNOWLEDGEMENTS
 
This work is supported by grants from Institut Pasteur and INSERM. G. E. D. is a Ph.D. student from Université Paris 7 and is supported by an Allocation du Ministère Français de la Recherche et de l’Enseignement Supérieur and by a fellowship from Association pour la recherche sur le cancer (ARC). R. G. is an assistant professor supported by Université Paris 7. There is no financial conflict of interest for this study. The B10BR Rag2/{gamma}c–/– mice breeding pairs and the FVB RAG2GFP transgenic mice breeding pairs are kind gifts of J. P. Disanto (Institut Pasteur, France) and M. C. Nussenzweig (The Rockefeller University, New York, NY, USA), respectively. We acknowledge A. Louise for cell sorting and the flow cytometry core facility (Institut Pasteur). We appreciate critical reading of the manuscript from P. Pereira.

Received November 14, 2007; revised January 7, 2008; accepted January 15, 2008.


arrow
REFERENCES
 
    1
  1. Godin, I., Cumano, A. (2002) The hare and the tortoise: an embryonic haematopoietic race Nat. Rev. Immunol. 2,593-604[Medline]
    2. 2
  2. Okada, S., Nakauchi, H., Nagayoshi, K., Nishikawa, S., Miura, Y., Suda, T. (1992) In vivo and in vitro stem cell function of c-kit- and Sca-1-positive murine hematopoietic cells Blood 80,3044-3050[Abstract/Free Full Text]
    3. 3
  3. Adolfsson, J., Mansson, R., Buza-Vidas, N., Hultquist, A., Liuba, K., Jensen, C. T., Bryder, D., Yang, L., Borge, O. J., Thoren, L. A., Anderson, K., Sitnicka, E., Sasaki, Y., Sigvardsson, M., Jacobsen, S. E. (2005) Identification of Flt3+ lympho-myeloid stem cells lacking erythro-megakaryocytic potential a revised road map for adult blood lineage commitment Cell 121,295-306[CrossRef][Medline]
    4. 4
  4. Forsberg, E. C., Serwold, T., Kogan, S., Weissman, I. L., Passegue, E. (2006) New evidence supporting megakaryocyte-erythrocyte potential of flk2/flt3+ multipotent hematopoietic progenitors Cell 126,415-426[CrossRef][Medline]
    5. 5
  5. Igarashi, H., Gregory, S. C., Yokota, T., Sakaguchi, N., Kincade, P. W. (2002) Transcription from the RAG1 locus marks the earliest lymphocyte progenitors in bone marrow Immunity 17,117-130[CrossRef][Medline]
    6. 6
  6. Adolfsson, J., Borge, O. J., Bryder, D., Theilgaard-Monch, K., Astrand-Grundstrom, I., Sitnicka, E., Sasaki, Y., Jacobsen, S. E. (2001) Upregulation of Flt3 expression within the bone marrow Lin(–)Sca1(+)c-kit(+) stem cell compartment is accompanied by loss of self-renewal capacity Immunity 15,659-669[CrossRef][Medline]
    7. 7
  7. Lai, A. Y., Kondo, M. (2006) Asymmetrical lymphoid and myeloid lineage commitment in multipotent hematopoietic progenitors J. Exp. Med. 203,1867-1873[Abstract/Free Full Text]
    8. 8
  8. Wagers, A. J., Weissman, I. L. (2006) Differential expression of {alpha}2 integrin separates long-term and short-term reconstituting Lin–/loThy1.1(lo)c-kit+ Sca-1+ hematopoietic stem cells Stem Cells 24,1087-1094[Abstract/Free Full Text]
    9. 9
  9. Akashi, K., Traver, D., Miyamoto, T., Weissman, I. L. (2000) A clonogenic common myeloid progenitor that gives rise to all myeloid lineages Nature 404,193-197[CrossRef][Medline]
    10. 10
  10. Kondo, M., Weissman, I. L., Akashi, K. (1997) Identification of clonogenic common lymphoid progenitors in mouse bone marrow Cell 91,661-672[CrossRef][Medline]
    11. 11
  11. Mebius, R. E., Miyamoto, T., Christensen, J., Domen, J., Cupedo, T., Weissman, I. L., Akashi, K. (2001) The fetal liver counterpart of adult common lymphoid progenitors gives rise to all lymphoid lineages, CD45+CD4+CD3– cells, as well as macrophages J. Immunol. 166,6593-6601[Abstract/Free Full Text]
    12. 12
  12. Traver, D., Miyamoto, T., Christensen, J., Iwasaki-Arai, J., Akashi, K., Weissman, I. L. (2001) Fetal liver myelopoiesis occurs through distinct, prospectively isolatable progenitor subsets Blood 98,627-635[Abstract/Free Full Text]
    13. 13
  13. Yokota, T., Kouro, T., Hirose, J., Igarashi, H., Garrett, K. P., Gregory, S. C., Sakaguchi, N., Owen, J. J., Kincade, P. W. (2003) Unique properties of fetal lymphoid progenitors identified according to RAG1 gene expression Immunity 19,365-375[CrossRef][Medline]
    14. 14
  14. Pelayo, R., Welner, R., Perry, S. S., Huang, J., Baba, Y., Yokota, T., Kincade, P. W. (2005) Lymphoid progenitors and primary routes to becoming cells of the immune system Curr. Opin. Immunol. 17,100-107[CrossRef][Medline]
    15. 15
  15. Cumano, A., Paige, C. J., Iscove, N. N., Brady, G. (1992) Bipotential precursors of B cells and macrophages in murine fetal liver Nature 356,612-615[CrossRef][Medline]
    16. 16
  16. Douagi, I., Andre, I. I., Ferraz, J. C., Cumano, A. (2000) Characterization of T cell precursor activity in the murine fetal thymus: evidence for an input of T cell precursors between days 12 and 14 of gestation Eur. J. Immunol. 30,2201-2210[Medline]
    17. 17
  17. Morrison, S. J., Hemmati, H. D., Wandycz, A. M., Weissman, I. L. (1995) The purification and characterization of fetal liver hematopoietic stem cells Proc. Natl. Acad. Sci. USA 92,10302-10306[Abstract/Free Full Text]
    18. 18
  18. Mebius, R. E., Rennert, P., Weissman, I. L. (1997) Developing lymph nodes collect CD4+CD3– LTβ+ cells that can differentiate to APC, NK cells, and follicular cells but not T or B cells Immunity 7,493-504[CrossRef][Medline]
    19. 19
  19. Martin, C. H., Aifantis, I., Scimone, M. L., von Andrian, U. H., Reizis, B., von Boehmer, H., Gounari, F. (2003) Efficient thymic immigration of B220+ lymphoid-restricted bone marrow cells with T precursor potential Nat. Immunol. 4,866-873[CrossRef][Medline]
    20. 20
  20. Wolber, F. M., Leonard, E., Michael, S., Orschell-Traycoff, C., Yoder, M., Srour, E. (2002) Roles of spleen and liver in development of the murine hematopoietic system Exp. Hematol. 30,1010-1019[CrossRef][Medline]
    21. 21
  21. Christensen, J. L., Wright, D. E., Wagers, A. J., Weissman, I. L. (2004) Circulation and chemotaxis of fetal hematopoietic stem cells PLoS Biol. 2,E75[CrossRef][Medline]
    22. 22
  22. Godin, I., Garcia-Porrero, J. A., Dieterlen-Lievre, F., Cumano, A. (1999) Stem cell emergence and hemopoietic activity are incompatible in mouse intraembryonic sites J. Exp. Med. 190,43-52[Abstract/Free Full Text]
    23. 23
  23. Bertrand, J. Y., Desanti, G. E., Lo-Man, R., Leclerc, C., Cumano, A., Golub, R. (2006) Fetal spleen stroma drives macrophage commitment Development 133,3619-3628[Abstract/Free Full Text]
    24. 24
  24. Mebius, R. E., Kraal, G. (2005) Structure and function of the spleen Nat. Rev. Immunol. 5,606-616[CrossRef][Medline]
    25. 25
  25. Brendolan, A., Ferretti, E., Salsi, V., Moses, K., Quaggin, S., Blasi, F., Cleary, M. L., Selleri, L. (2005) A Pbx1-dependent genetic and transcriptional network regulates spleen ontogeny Development 132,3113-3126[Abstract/Free Full Text]
    26. 26
  26. Carlyle, J. R., Zuniga-Pflucker, J. C. (1998) Requirement for the thymus in {alpha}β T lymphocyte lineage commitment Immunity 9,187-197[CrossRef][Medline]
    27. 27
  27. Kiel, M. J., Iwashita, T., Yilmaz, O. H., Morrison, S. J. (2005) Spatial differences in hematopoiesis but not in stem cells indicate a lack of regional patterning in definitive hematopoietic stem cells Dev. Biol. 283,29-39[CrossRef][Medline]
    28. 28
  28. Colucci, F., Turner, M., Schweighoffer, E., Guy-Grand, D., Di Bartolo, V., Salcedo, M., Tybulewicz, V. L., Di Santo, J. P. (1999) Redundant role of the Syk protein tyrosine kinase in mouse NK cell differentiation J. Immunol. 163,1769-1774[Abstract/Free Full Text]
    29. 29
  29. Yu, W., Nagaoka, H., Jankovic, M., Misulovin, Z., Suh, H., Rolink, A., Melchers, F., Meffre, E., Nussenzweig, M. C. (1999) Continued RAG expression in late stages of B cell development and no apparent re-induction after immunization Nature 400,682-687[CrossRef][Medline]
    30. 30
  30. Desanti, G. E., Bertrand, J. Y., Golub, R. (2007) Fetal spleen development, the ride toward multiple functions Functional Development and Embryology 1,78-90
    31. 31
  31. Borghesi, L., Hsu, L. Y., Miller, J. P., Anderson, M., Herzenberg, L., Herzenberg, L., Schlissel, M. S., Allman, D., Gerstein, R. M. (2004) B lineage-specific regulation of V(D)J recombinase activity is established in common lymphoid progenitors J. Exp. Med. 199,491-502[Abstract/Free Full Text]
    32. 32
  32. Payne, K. J., Medina, K. L., Kincade, P. W. (1999) Loss of c-kit accompanies B-lineage commitment and acquisition of CD45R by most murine B-lymphocyte precursors Blood 94,713-723[Abstract/Free Full Text]
    33. 33
  33. Yu, W., Misulovin, Z., Suh, H., Hardy, R. R., Jankovic, M., Yannoutsos, N., Nussenzweig, M. C. (1999) Coordinate regulation of RAG1 and RAG2 by cell type-specific DNA elements 5' of RAG2 Science 285,1080-1084[Abstract/Free Full Text]
    34. 34
  34. Yokoyama, W. M., Kim, S., French, A. R. (2004) The dynamic life of natural killer cells Annu. Rev. Immunol. 22,405-429[CrossRef][Medline]
    35. 35
  35. Velardi, A., Cooper, M. D. (1984) An immunofluorescence analysis of the ontogeny of myeloid, T, and B lineage cells in mouse hemopoietic tissues J. Immunol. 133,672-677[Abstract]
    36. 36
  36. Martin, F., Kearney, J. F. (2002) Marginal-zone B cells Nat. Rev. Immunol. 2,323-335[CrossRef][Medline]
    37. 37
  37. Wardemann, H., Boehm, T., Dear, N., Carsetti, R. (2002) B-1a B cells that link the innate and adaptive immune responses are lacking in the absence of the spleen J. Exp. Med. 195,771-780[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental figures and tables
Right arrow All Versions of this Article:
jlb.1107755v1
83/5/1145    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Desanti, G. E.
Right arrow Articles by Golub, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Desanti, G. E.
Right arrow Articles by Golub, R.