Osteoblasts support B-lymphocyte commitment and differentiation from hematopoietic stem cells

Early B lymphopoiesis in mammals is induced within the bone marrow (BM) microenvironment, but which cells constitute this niche is not known. Previous studies had shown that osteoblasts (OBs) support hematopoietic stem cell (HSC) proliferation and myeloid differentiation. We now find that purified primary murine OBs also support the differentiation of primitive hematopoietic stem cells through lymphoid commitment and subsequent differentiation to all stages of B-cell precursors and mature B cells. Lin(-)Sca-1(+)Rag-2(-) BM cell differentiation to B cells requires their attachment to OBs in vitro, and this developmental process is mediated via VCAM-1, SDF-1, and IL-7 signaling induced by parathyroid hormone (PTH). Addition of cytokines produced by nonosteoblastic stromal cells (c-Kit ligand, IL-6, and IL-3) shifted the cultures toward myelopoiesis. Confirming the role of OBs in B lymphopoiesis, we found that selective elimination of osteoblasts in Col2.3Delta-TK transgenic mice severely depleted pre-pro-B and pro-B cells from BM, preceding any decline in HSCs. Taken together, these results demonstrate that osteoblasts are both necessary and sufficient for murine B-cell commitment and maturation, and thereby constitute the cellular homolog of the avian bursa of Fabricius.     

The flt3 ligand supports proliferation of lymphohematopoietic progenitors and early B-lymphoid progenitors

We have examined the effects of the murine ligand (FL) for the flt3/flk2 tyrosine kinase receptor on the proliferation of murine lymphohematopoietic progenitors as well as committed myeloid and B-cell progenitors. In the presence of erythropoietin, FL alone supported scant colony formation from enriched marrow cells of normal mice. However, when it was combined with interleukin-3 (IL-3), steel factor (SF), or IL-11, FL significantly enhanced colony formation. When tested on enriched marrow cells from 5-fluorouracil (5-FU)-treated mice, FL neither enhanced IL-3-dependent colony formation nor synergized with SF in support of colony formation. However, FL synergized with IL-6, IL- 11, or granulocyte-colony stimulating factor (G-CSF) in support of formation of various types of colonies, including multilineage colonies. Approximately 30% of these colonies yielded pre-B-cell colonies when replated in secondary cultures containing SF and IL-7, indicating that 2-cytokine combinations, including FL and IL-6, IL-11, or G-CSF can support the proliferation of primitive lymphohematopoietic progenitors. FL, by itself and in synergy with IL-7 or SF, supported the proliferation of B-cell progenitors. These results show that FL has a wide range of activities in early hematopoiesis and B lymphopoiesis.     

Inhibition of CD10/neutral endopeptidase 24.11 promotes B-cell reconstitution and maturation in vivo.

The common acute lymphoblastic leukemia antigen [(CALLA) CD10, neutral endopeptidase 24.11 (NEP)] is a cell-surface zinc metalloprotease expressed by a subpopulation of early murine B-lymphoid progenitors and by bone marrow stromal cells that support the earliest stages of B lymphopoiesis. In previous in vitro studies in which uncommitted murine hematopoietic progenitors plated on a stromal cell layer differentiate into immature B cells, the inhibition of CD10/NEP increased early lymphoid colony numbers. To further characterize CD10/NEP function during lymphoid ontogeny in vivo, we utilized a Ly5 congenic mouse model in which the lymphoid differentiation of uncommitted hematopoietic progenitors from Ly5.1 donors was followed in sublethally irradiated Ly5.2 recipients treated with a specific long-acting CD10/NEP inhibitor (N-[L-(1-carboxy-2-phenyl)ethyl]-L-phenylalanyl-beta- alanine (SCH32615)). The expression of Ly5.1, B220, and surface IgM (sIgM) was utilized to characterize donor-derived hematopoietic cells (Ly5.1+), B lymphocytes (B220+), and mature B cells (B220+ sIgM+) from the lymphoid organs of recipient animals treated with SCH32615 or vehicle alone. SCH32615-treated animals had higher percentages of Ly5.1+ donor splenocytes than animals treated with vehicle alone (16.9% vs. 10.4%, 63% increase, P = 0.013). Animals treated with the CD10/NEP inhibitor also had relatively more Ly5.1+ splenic B (B220+) cells than vehicle-treated animals (14.4% vs. 8.2%, 75% increase, P = 0.018). To more specifically characterize the effects of CD10/NEP inhibition on B-cell differentiation, Ly5.1+ splenocytes from animals treated with SCH32615 or vehicle alone were analyzed for coexpression of B220 and sIgM. Animals treated with the CD10/NEP inhibitor had a significantly higher percentage of mature donor B cells (Ly5.1+ B220+ sIgM+, 10.2% vs. 5.2%, 90% increase, P = 0.006) and a more modest relative increase in immature donor B cells (Ly5.1+ B220+ sIgM-, 4.7% vs. 3.4%, 38% increase, P = not significant). Taken together, these results suggest that CD10/NEP inhibition promotes the reconstitution and maturation of splenic B cells. Therefore, CD10/NEP may function to regulate B-cell ontogeny in vivo by hydrolyzing a peptide substrate that stimulates B-cell proliferation and/or differentiation.     

Native associations of early hematopoietic stem cells and stromal cells isolated in bone marrow cell aggregates

In suspensions of murine bone marrow, many stromal cells are tightly entwined with hematopoietic cells. These cellular aggregations appear to exist normally within the marrow. Previous studies showed that lymphocytes and stem cells adhered to stromal cells via vascular cell adhesion molecule 1 (VCAM1). Injection of anti-VCAM1 antibody into mice disrupts the aggregates, showing the importance of VCAM1 in the adhesion between stromal cells and hematopoietic cells in vivo. Early hematopoietic stem cells were shown to be enriched in aggregates by using a limiting-dilution culture assay. Myeloid progenitors responsive to WEHI-3CM in combination with stem cell factor (c-kit ligand) and B220- B-cell progenitors responsive to insulin-like growth factor-1 in combination with interleukin-7 are not enriched. We propose a scheme of stromal cell-hematopoietic cell interactions based on the cell types selectively retained within the aggregates. The existence of these aggregates as native elements of bone marrow organization presents a novel means to study in vivo stem cell-stromal cell interaction.     

In vitro requirement for periostin in B lymphopoiesis

B lymphopoiesis arrests in rabbits by 4 months of age. To identify molecules that contribute to this arrest, cDNA-representational difference analysis on BM stromal cells from young and adult rabbits showed that expression of Postn that encodes for the extracellular matrix protein periostin dramatically reduced with age. Postn-small interfering RNA OP9 cells lost their capacity to support B-cell development from rabbit or murine BM cells, and reexpression of periostin restored this potential, indicating an in vitro requirement for periostin in B lymphopoiesis. In our system, we determined that periostin deficiency leads to increased cell death and decreased proliferation of B-lineage progenitors. Further, RGD peptide inhibition of periostin/α(v)β(3) interaction resulted in a marked decrease in B lymphopoiesis in vitro. Microarray analysis of the Postn-small interfering RNA OP9 cells showed decreased expression of key B-lymphopoietic factors, including IL-7 and CXCL12. In vivo, unidentified molecule(s) probably compensate periostin loss because Postn(-/-) mice had normal numbers of B-cell progenitors in BM. We conclude that the decline in periostin expression in adult rabbit BM does not solely explain the arrest of B lymphopoiesis. However, the interaction of periostin with α(v)β(3) on lymphoid progenitors probably provides both proliferative and survival signals for cells in the B-cell development pathway.     

Differential effects of dehydroepiandrosterone (DHEA) on murine lymphopoiesis and myelopoiesis.

Dehydroepiandrosterone (DHEA) inhibits murine lymphopoiesis, especially after sublethal irradiation, without affecting mature lymphocyte function. We demonstrate here that dietary DHEA differentially affected myelopoiesis and lymphopoiesis in sublethally irradiated mice. Erythropoietic progenitor cell and stem cell function was not affected by DHEA although the magnitude of granulopoiesis was slightly reduced. Regeneration of marrow B220+ B cells, natural killer (NK) function, and thymus repopulation were significantly delayed in DHEA-fed mice. Interleukin 2 (IL-2) failed to restore NK activity of DHEA-fed mice to normal levels. Marrow from DHEA-fed mice contained competent thymocyte progenitors capable of repopulating thymi of irradiated hosts fed a control diet but not those fed the DHEA diet. Thus DHEA inhibits lymphopoiesis while sparing myelopoiesis, affecting the growth and maturation of T, B, and NK cells by a mechanism other than the inhibition of IL-2 production.     

Oncostatin M suppresses generation of lymphoid progenitors in fetal liver by inhibiting the hepatic microenvironment

OBJECTIVE: Interaction between hematopoietic cells and stromal cells is important for regulation of hematopoiesis. Numerous soluble and membrane-bound factors directly regulating hematopoiesis have been documented, but little is known about how stromal cell activity is controlled. We previously reported that fetal hepatic cells in primary culture create the hematopoietic microenvironment and support expansion of blood cells from hematopoietic stem cells. In this study, we focused on lymphopoiesis reconstituted in our culture system and analyzed how stroma-mediated lymphopoiesis is regulated during embryonic development. MATERIALS AND METHODS: Subconfluent cultures of murine fetal hepatic cells were cocultured with hematopoietic stem cells derived from fetal liver in the presence of various cytokines. After 10 days of incubation, hematopoietic cells floating over the stromal layer were analyzed by various assays, including cell proliferation and FACS analysis. RESULTS: We found that oncostatin M, an inducer of hepatic development, strongly inhibited generation of B220(+) lymphocytic cells and colony-forming unit-interleukin-7 (CFU-IL-7) from hematopoietic stem cells in our coculture system. In contrast, oncostatin M did not directly inhibit proliferation of B cells in response to IL-7 and SCF in semisolid cultures. Analysis of antigen expression in lymphoid cells revealed that oncostatin M apparently did not arrest cells at a particular stage of B-cell development. CONCLUSIONS: The results suggest that oncostatin M inhibits lymphopoiesis by suppressing stromal activity of fetal hepatic cells to stimulate generation of CFU-IL-7 from their progenitors rather than by acting directly on lymphocytic cells.     

FLT3 ligand and not TSLP is the key regulator of IL-7-independent B-1 and B-2 B lymphopoiesis

Phenotypically and functionally distinct progenitors and developmental pathways have been proposed to exist for fetally derived B-1 and conventional B-2 cells. Although IL-7 appears to be the primary cytokine regulator of fetal and adult B lymphopoiesis in mice, considerable fetal B lymphopoiesis and postnatal B cells are sustained in the absence of IL-7; in humans, B-cell generation is suggested to be largely IL-7-independent, as severe combined immune-deficient patients with IL-7 deficiency appear to have normal B-cell numbers. However, the role of other cytokines in IL-7-independent B lymphopoiesis remains to be established. Although thymic stromal lymphopoietin (TSLP) has been proposed to be the main factor driving IL-7-independent B lymphopoiesis and to distinguish fetal from adult B-cell progenitor development in mice, recent studies failed to support a primary role of TSLP in IL-7-independent fetal B-cell development. However, the role of TSLP in IL-7-independent adult B lymphopoiesis and in particular in regulation of B-1 cells remains to be established. Here we demonstrate that, rather than TSLP, IL-7 and FLT3 ligand are combined responsible for all B-cell generation in mice, including recently identified B-1-specified cell progenitors. Thus, the same IL-7- and FLT3 ligand-mediated signal-ing regulates alternative pathways of fetal and adult B-1 and B-2 lymphopoiesis.     

Stromal cells in myeloid and lymphoid long-term bone marrow cultures can support multiple hemopoietic lineages and modulate their production of hemopoietic growth factors

Hemopoiesis in long-term bone marrow cultures (LTBMC) is dependent on adherent stromal cells that form an in vitro hemopoietic microenvironment. Myeloid bone marrow cultures (MBMC) are optimal for myelopoiesis, while lymphoid bone marrow cultures (LBMC) only support B lymphopoiesis. The experiments reported here have made a comparative analysis of the two cultures to determine whether the stromal cells that establish in vitro are restricted to the support of myelopoiesis or lymphopoiesis, respectively, and to examine how the different culture conditions affect stromal cell physiology. In order to facilitate this analysis, purified populations of MBMC and LBMC stroma were prepared by treating the LTBMC with the antibiotic mycophenolic acid; this results in the elimination of hemopoietic cells while retaining purified populations of functional stroma. Stromal cell cultures prepared and maintained under MBMC conditions secreted myeloid growth factors that stimulated the growth of granulocyte-macrophage colonies, while no such activity was detected from purified LBMC stromal cultures. However, this was not due to the inability of LBMC stroma to mediate this function. Transfer of LBMC stromal cultures to MBMC conditions resulted in an induction of myeloid growth factor secretion. When seeded under these conditions with stromal cell- depleted populations of hemopoietic cells, obtained by passing marrow through nylon wool columns, the LBMC stromal cells could support long- term myelopoiesis. Conversely, transfer of MBMC stroma to LBMC conditions resulted in a cessation of myeloid growth factor secretion; on seeding these cultures with nylon wool-passed marrow, B lymphopoiesis, but not myelopoiesis, initiated. These findings indicate that the stroma in the different LTBMC are not restricted in their hemopoietic support capacity but are sensitive to culture conditions in a manner that may affect the type of microenvironment formed.     

B-lymphopoiesis is stopped by mobilizing doses of G-CSF and is rescued by overexpression of the anti-apoptotic protein Bcl2

Osteoblasts are necessary to B lymphopoiesis and mobilizing doses of G-CSF or cyclophosphamide inhibit osteoblasts, whereas AMD3100/Plerixafor does not. However, the effect of these mobilizing agents on B lymphopoiesis has not been reported. Mice (wild-type, knocked-out for TNF-α and TRAIL, or over-expressing Bcl-2) were mobilized with G-CSF, cyclophosphamide, or AMD3100. Bone marrow, blood, spleen and lymph node content in B cells was measured. G-CSF stopped medullar B lymphopoiesis with concomitant loss of B-cell colony-forming units, pre-pro-B, pro-B, pre-B and mature B cells and increased B-cell apoptosis by an indirect mechanism. Overexpression of the anti-apoptotic protein Bcl2 in transgenic mice rescued B-cell colony forming units and pre-pro-B cells in the marrow, and prevented loss of all B cells in marrow, blood and spleen. Blockade of endogenous soluble TNF-α with Etanercept, or combined deletion of the TNF-α and TRAIL genes did not prevent B lymphopoiesis arrest in response to G-CSF. Unlike G-CSF, treatments with cyclophosphamide or AMD3100 did not suppress B lymphopoiesis but caused instead robust B-cell mobilization. G-CSF, cyclophosphamide and AMD3100 have distinct effects on B lymphopoiesis and B-cell mobilization with: 1) G-CSF inhibiting medullar B lymphopoiesis without mobilizing B cells in a mechanism distinct from the TNF-α-mediated loss of B lymphopoiesis observed during inflammation or viral infections; 2) CYP mobilizing B cells but blocking their maturation; and 3) AMD3100 mobilizing B cells without affecting B lymphopoiesis. These results suggest that blood mobilized with these three agents may have distinct immune properties.     

Expression of granulocyte colony-stimulating factor and granulocyte colony-stimulating factor receptor genes in partially overlapping monoclonal B-cell populations from chronic lymphocytic leukemia patients

B lymphocytes were purified from the peripheral blood of 30 B-cell chronic lymphocytic leukemia (B-CLL) patients and tested for the ability to produce granulocyte colony-stimulating factor (G-CSF) in vitro. Fifteen Staphylococcus aureus Cowan I (SAC)-stimulated, but not unstimulated, B-cell suspensions produced G-CSF in short-term cultures. Accordingly, G-CSF mRNA was detected only in SAC-stimulated B cells. Five CLL B-cell fractions that released G-CSF following exposure to SAC were also incubated with CD40 or anti-mu antibodies in the presence or absence of recombinant (r) interleukin-2 (IL-2) or IL-4. The 5 cell suspensions produced G-CSF only on culture with CD40 monoclonal antibody in combination with rIL-2 or rIL-4. CD5+ B lymphocytes, which represent the normal counterparts of most B-CLL proliferations, did not produce G-CSF under any of the above culture conditions. G-CSF produced by leukemic B lymphocytes was biologically active, because conditioned media of SAC-stimulated cells supported the in vitro growth of myeloid colonies from normal bone marrow progenitors. The colony stimulating activity of CLL B-cell supernatants was ascribed to both G-CSF and granulocyte-macrophage colony stimulating factor. G-CSF receptors (G- CSFRs) were detected on freshly isolated B lymphocytes from 7 of 11 B- CLL patients; 5 of these cell suspensions produced G-CSF in culture, whereas 2 did not. rG-CSF rescued 3 of the 7 G-CSFR+ cell fractions from spontaneous apoptosis but had no effect on their in vitro proliferation.     

Regulation of cell survival during B lymphopoiesis in mouse bone marrow: enhanced pre-B-cell apoptosis in CSF-1-deficient op/op mutant mice

OBJECTIVE: Osteopetrotic (op/op) mice are deficient in macrophages and osteoclasts due to a CSF-1 gene mutation. The aim of this study was to evaluate the effect of these deficiencies and of CSF-1-dependent mechanisms on B lymphopoiesis in bone marrow, with special reference to the apoptotic activity of precursor B cells. MATERIALS AND METHODS: B-cell development and apoptosis were examined in the bone marrow of op/op mice using immunofluorescence labeling and flow cytometry. Short-term cultures of bone marrow were used to evaluate the effect of recombinant CSF-1 on the rate of B-cell apoptosis. RESULTS: Bone marrow cellularity was greatly reduced in op/op mice compared with normal littermates. However, precursor B cells were disproportionately decreased, most markedly at the pre-B-cell stage. Precursor B cells, particularly pre-B cells, displayed elevated apoptotic incidences both ex vivo and in short-term culture. Addition of recombinant CSF-1 reduced the incidence of apoptosis among precursor B cells in short-term cultures of whole bone marrow suspensions from normal mice but not in cultures of sorted B220+ B-lineage cells. CONCLUSIONS: The finding of increased pre-B-cell apoptosis in op/op mice provides evidence that CSF-1-dependent mechanisms can strongly influence the survival of precursor B cells in mouse bone marrow, particularly at the pro-B/pre-B cell transition. It is proposed that the local or systemic levels of CSF-1 during ontogeny may thus play a role in regulating B-cell production within the bone marrow microenvironment.     

Support of early B-cell differentiation in mouse fetal liver by stromal cells and interleukin-7.

We compared the development of B-cell progenitors with that of myeloid progenitors in fetal liver cells at various gestational ages. Day 12 to 14 fetal liver cells did not form pre-B-cell colonies. Pre-B-cell colonies were developed from day 15 fetal liver cells. The incidence of colonies increased with increases in gestational age and reached a maximum on days 18 to 19. In contrast, the incidence of myeloid colonies formed in the presence of interleukin-3 (IL-3) and erythropoietin did not change significantly during days 13 to 21 of gestation. After coculturing day 13 fetal liver cells with IL-7-producing stromal cell line ST-2, they could respond to IL-7 and proliferate. Analysis of the phenotypes showed that day 13 fetal liver cells were B220-, IgM-, while culturing day 13 fetal liver cells with ST-2 and untreated day 18 fetal liver cells contained the population of B220+ cells. Even in the presence of IL-7-defective stromal cell line FLS-3, IL-7-responsive cells could be induced from day 13 fetal liver cells. IL-7 acted on B220+ cells and induced pre-B-cell colonies that contained IgM+ cells in the methylcellulose culture. IL-7 mRNA was expressed in days 13 and 18 fetal liver cells but not in pre-B cells or adult liver cells. From these findings, it is suggested that stromal cells or stromal-derived factors but not IL-7 were required for the differentiation from B220- cells to B220+ cells. In the second stage, B220+, IgM- cells proliferated and some of them differentiated to IgM+ cells in the presence of IL-7 alone. The two-step model can apply to in vivo early B lymphopoiesis.     

In vitro generation of hematopoietic stem cells from an embryonic stem cell line.

Hematopoietic stem cells (HSC) are unique in that they give rise both to new stem cells (self-renewal) and to all blood cell types. The cellular and molecular events responsible for the formation of HSC remain unknown mainly because no system exists to study it. Embryonic stem (ES) cells were induced to differentiate by coculture with the stromal cell line RP010 and the combination of interleukin (IL) 3, IL-6, and F (cell-free supernatants from cultures of the FLS4.1 fetal liver stromal cell line). Cell cytometry analysis of the mononuclear cells produced in the cultures was consistent with the presence of PgP-1+ Lin- early hematopoietic (B-220- Mac-1- JORO 75- TER 119-) cells and of fewer B-220+ IgM- B-cell progenitors and JORO 75+ T-lymphocyte progenitors. The cell-sorter-purified PgP-1+ Lin- cells produced by induced ES cells could repopulate the lymphoid, myeloid, and erythroid lineages of irradiated mice. The ES-derived PgP-1+ Lin- cells must possess extensive self-renewal potential, as they were able to produce hematopoietic repopulation of secondary mice recipients. Indeed, marrow cells from irradiated mice reconstituted (15-18 weeks before) with PgP-1+ Lin- cell-sorter-purified cells generated by induced ES cells repopulated the lymphoid, myeloid, and erythroid lineages of secondary mouse recipients assessed 16-20 weeks after their transfer into irradiated secondary mice. The results show that the culture conditions described here support differentiation of ES cells into hematopoietic cells with functional properties of HSC. It should now be possible to unravel the molecular events leading to the formation of HSC.     

B220- bone marrow progenitor cells from New Zealand black autoimmune mice exhibit an age-associated decline in Pre-B and B-cell generation

New Zealand Black (NZB) autoimmune mice exhibit progressive, age- dependent reduction in bone marrow pre-B cells. To ascertain the capacity of NZB bone marrow B220- cells to generate pre-B cells in a supportive environment, B-lineage (B220+) cell-depleted and T-cell- depleted bone marrow cells from NZB mice at 1 to 3, 6, and 10 to 11 months of age were adoptively transferred into irradiated (200R) C.B17 severe combined immunodeficient (SCID) mice. Bone marrow pre-B cells (sIgM- CD43[S7]- B220+) were assessed 3 and 10 weeks posttransfer. Pre- B cells and B cells were reconstituted in SCID recipients of older NZB progenitor cells by 10 weeks posttransplant, in contrast to the very low numbers of pre-B cells present in the donor bone marrow. However, B220- bone marrow progenitor cells from greater than 10-month-old NZB donors were deficient in the reconstitution of both pre-B and B cells in SCID recipients at 3 weeks post-transfer. This reflected a slower kinetics of repopulation, because older NZB-->SCID recipients had numbers of both pre-B and B cells similar to recipients of young NZB progenitor cells by 10 weeks posttransplant. Adoptive transfer of equal mixtures of BALB/c and older NZB bone marrow B220- progenitor cells into irradiated C.B17 SCID recipients failed to demonstrate active suppression. These results suggest that, with age, NZB bone marrow has reduced numbers and/or function of early B220- B-lineage progenitors. Consistent with this hypothesis, B220- bone marrow cells from older NZB mice were deficient in progenitors capable of yielding interleukin-7 (IL-7) responsive pre-B cells in vitro on stimulation with the pre-B- cell potentiating factor, insulin-like growth factor 1 (IGF-1).     

Crucial role of FLT3 ligand in immune reconstitution after bone marrow transplantation and high-dose chemotherapy

Almost 5 decades after the first clinical transplantations, delayed immune reconstitution remains a considerable hurdle in bone marrow transplantation, and the mechanisms regulating immune reconstitution after transplantation remain to be established. Whereas adult fms-like tyrosine kinase 3 ligand-deficient (FL(-/-)) mice have reduced numbers of early B- and T-cell progenitors, they sustain close to normal levels of mature B and T cells. Herein, we demonstrate that adult bone marrow cells fail to reconstitute B-cell progenitors and conventional B cells in lethally irradiated FL(-/-) recipients, which also display delayed kinetics of T-cell reconstitution. Similarly, FL is essential for B-cell regeneration after chemotherapy-induced myeloablation. In contrast, fetal progenitors reconstitute B lymphopoiesis in FL(-/-) mice, albeit at reduced levels. A critical role of FL in adult B lymphopoiesis is further substantiated by an age-progressive decline in peripheral conventional B cells in FL(-/-) mice, whereas fetally and early postnatally derived B1 and marginal zone B cells are sustained in a FL-independent manner. Thus, FL plays a crucial role in sustaining conventional B lymphopoiesis in adult mice and, as a consequence, our findings implicate a critical role of FL in promoting immune reconstitution after myeloablation and bone marrow transplantation.     

A cell-autonomous requirement for CXCR4 in long-term lymphoid and myeloid reconstitution

Mice lacking the chemokine stromal cell-derived factor/pre-B cell growth stimulating factor or its primary physiological receptor CXCR4 revealed defects in B lymphopoiesis and bone marrow myelopoiesis during embryogenesis. We show here that adoptive transfer experiments reveal a deficiency in long-term lymphoid and myeloid repopulation in adult bone marrow by CXCR4-/- fetal liver cells, although stromal cell-derived factor/pre-B cell growth stimulating factor-/- fetal liver cells yield normal multilineage reconstitution. These findings indicate that CXCR4 is required cell autonomously for lymphoid and myeloid repopulation in bone marrow. In addition, CXCR4-/- fetal liver cells generated much more severely reduced numbers of B cells relative to other lineages in bone marrow. Furthermore, the repopulation of c-kit+ Sca-1(+) linlow/- cells by CXCR4-/- fetal liver cells was less affected compared with c-kit+ Sca-1(-) linlow/- cells. By previous studies, it has been shown that c-kit+ Sca-1(+) linlow/- cells are highly purified primitive hematopoietic progenitors and that c-kit+ Sca-1(-) linlow/- cells are more committed hematopoietic progenitors in mice. Thus, CXCR4 may play an essential role in generation and/or expansion of early hematopoietic progenitors within bone marrow.