Mouse Bone MarrowCharacterization of Pro-B Cells in Adult Phenotypic Distinction and Functional
William B. Slayton and Gerald J. SpangrudeJ. Elenitoba-Johnson, L. Jeanne Pierce, Anne Wiesmann, Mariluz P. Mojica, S. Scott Perry, A. Elena Searles, Kojo S.
Phenotypic Distinction and Functional Characterization ofPro-B Cells in Adult Mouse Bone Marrow1
Mariluz P. Mojica,* S. Scott Perry,† A. Elena Searles,‡ Kojo S. J. Elenitoba-Johnson,†
L. Jeanne Pierce,§ Anne Wiesmann,§ William B. Slayton,¶ and Gerald J. Spangrude2†‡§
A lymphoid-committed progenitor population was isolated from mouse bone marrow based on the cell surface phenotype Thy-1.1negSca-1posc-Kit lowLin neg. These cells were CD43posCD24pos on isolation and proliferated in response to the cytokine combi-nation of steel factor, IL-7, and Flt3 ligand. Lymphoid-committed progenitors could be segregated into more primitive and moredifferentiated subsets based on expression of AA4.1. The more differentiated subset generated only B lymphoid cells in 92% of totalcolonies assayed, lacked T lineage potential, and expressed Pax5. These studies have therefore defined and isolated a B lymphoid-committed progenitor population at a developmental stage corresponding to the initial expression of CD45R.The Journal ofImmunology,2001, 166: 3042–3051.
Combinations of surface proteins expressed by cells at dif-ferent stages of blood cell development have enabled theisolation and subsequent functional characterization of
phenotypically defined hemopoietic progenitor cell populations. Inthe adult mouse, the bone marrow (BM)3 is the site of hemopoiesiswhere mature blood cell lineages are generated from self-renewingmultipotent hemopoietic stem cells (HSC) (1). These HSC expresslow levels of Thy-1.1 and high levels of stem cell Ag-1 (Sca-1pos)and lack expression of lineage-associated markers (Linneg) (2).Phenotypically defined progenitor cell populations with restrictedmyeloid and/or lymphoid lineage potentials have recently beendescribed (3–6). Expression of CD45R has been used by a numberof investigators to isolate and characterize lymphoid progenitors(7, 8), while stages of B lymphoid development before expressionof CD45R are only beginning to be explored (4, 9, 10).
A goal in enriching for early B lymphoid progenitors is to sep-arate these cells from both HSC and differentiated cells of the Band other blood cell lineages. The latter can be achieved by de-pleting BM of cells expressing CD45R and other lineage-associ-ated markers. Segregating early lymphoid progenitor cells fromHSC has been more challenging. Previous studies have shown thatalthough the Thy-1.1lowSca-1posLinneg (Thy-1.1low) cell popula-tion is markedly enriched for HSC, it is functionally heterogeneous
(11, 12). This heterogeneity, which is reflected in the ability of thecell population to mediate both short term and long term BM en-graftment, can be dissected using separations based on cell cycleactivity and activation state (13, 14). Although experiments de-signed to segregate progenitors for lymphoid lineages have shownthat subsets of Thy-1.1low HSC engraft in the BM, the kinetics ofB cell engraftment exhibit a 2-wk delay relative to whole BMtransplants (15). This suggests that a lymphoid progenitor cell pop-ulation that is present in normal BM is missing from the Thy-1.1low HSC population.
The c-Kit molecule has also been used extensively as a cellsurface marker in the purification of HSCs (16–18). We recentlyfound that the c-Kitpossubset of Sca-1posLinnegBM cells containstwo populations of cells which differ with respect to expression ofThy-1.1. The majority of Thy-1.1low cells express c-Kit at highlevels (c-Kithigh), while Sca-1posLinneg BM cells lacking the Thy-1.1 Ag (Thy-1.1negcells) include c-Kitlow as well as c-Kithigh sub-sets. Transplant studies demonstrated that the Thy-1.1lowc-Kithigh
subset mediated full hemopoietic engraftment of lethally irradiatedrecipient animals with prominent erythroid, myeloid and plateletreconstitution and delayed lymphoid engraftment (19). In contrast,the Thy-1.1neg subset failed to provide erythroid and platelet en-graftment to transplant recipients, but mediated rapid lymphoidengraftment with a minor degree of myeloid recovery. In the ab-sence of full hemopoietic recovery, transplant recipients of Thy-1.1neg cells survive only 25–30 days before death due to hemo-poietic failure.
The current studies were initiated to resolve the lymphoid and my-eloid potentials of Thy-1.1negcells at a clonal level. Using a numberof additional phenotypic markers, we demonstrate three separate pro-genitor populations within the Thy-1.1neg subset of Sca-1posLinneg
BM cells. These include separate committed progenitors for lymphoidand myeloid (predominantly macrophage) lineages as well as a mixedlineage progenitor population. Interestingly, expression of the AA4.1Ag defines the onset of Pax5 transcription and marks the loss of pro-Tcell potential. Thus, these studies have defined a very early stage ofmouse pro-B cell development.
Materials and MethodsMouse strains
B6.PL and AKR mice were obtained from The Jackson Laboratory (BarHarbor, ME), while C57BL/Ka, B6-Thy-1.1-Ly-5.1, and B6-Ly-1.1 mice
Departments of *Human Genetics,†Pathology,‡Medicine (Division of Hematology),§Oncological Sciences, and¶Pediatrics, University of Utah, Salt Lake City, UT 84132
Received for publication August 28, 2000. Accepted for publication December18, 2000.
The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by grants from the National Institutes of Health(RO1HL56857 and P50DK49219), the Lymphoma Foundation, and the Blood andMarrow Transplant Program at the University of Utah. The Flow Cytometry andIrradiation core facilities of the Huntsman Cancer Institute, supported by NationalCancer Institute Cancer Center Support Grant P30CA42014, were used for thesestudies.2 Address correspondence and reprint requests to Dr. Gerald J. Spangrude, Department ofOncological Sciences, University of Utah, 50 North Medical Drive, Room 5C334, SaltLake City, UT 84132. E-mail address: [email protected] Abbreviations used in this paper: BM, bone marrow; HSC, hemopoietic stem cells;STL, Steel factor; Flt3L, Flt3 ligand; Epo, erythropoietin; Linneg, bone marrow cellsdepleted of cells expressing any of a panel of lineage-specific Ags; Thy-1.1low, Sca-11c-Kit1LinnegThy-1.1low BM cells; Thy-1.1neg, Sca-11c-Kit1LinnegThy-1.1negBMcells; PI, propidium iodide; SAv, streptavidin; Rag, recombinase-activating gene.
were bred and maintained at the Animal Resource Center facility of theUniversity of Utah. Mice used were 4–16 wk of age.
Cytokines and Abs
Steel factor (STL) and G-CSF were gifts from Gemini Science (San Diego,CA), a subsidiary of Kirin Brewery (Tokyo, Japan). Flt3 ligand (Flt3L) andIL-6 were kindly provided by Immunex (Seattle, WA). Recombinant hu-man erythropoietin (Epo) was purchased from Ortho (Raritan, NJ). Re-combinant murine IL-3 and IL-7 were purchased from PeproTech (RockyHill, NJ). The cytokines were used at the following concentrations: STL,100 ng/ml; G-CSF, 10 ng/ml; Flt3L, 75 ng/ml; IL-6, 20 ng/ml; Epo, 5U/ml; IL-3, 10 ng/ml; and IL-7, 10 ng/ml.
mAbs against CD8 (53-6.7), Mac1 (M1/70), erythrocytes (TER119),Gr-1 (RB6-8C5), CD3 (KT3-1.1), CD5 (53-7.3), CD2 (Rm2.2), mouse Ig(RAM-HB58), CD45R (RA3-6B2), Thy-1.1 (19XE5), c-Kit (ACK-4),early B cell (AA4.1), and IL-7R were purified from media of culturedhybridoma cell lines, while the mAb against CD19 was purchased fromPharMingen (San Diego, CA). mAbs used for cell surface staining ofCD45R, Thy-1.1, c-Kit, Gr-1, CD62L, AA4.1, and IL-7R were conjugatedwith biotin, PE, or FITC in our laboratory. Biotinylated Abs were second-arily stained with either PE-streptavidin (PE-SAv; Biomeda, Foster City,CA) or PE-Texas Red-SAv (Red613; Life Technologies, Grand Island,NY). In addition, PE-conjugated mAbs to Sca-1, Mac-1, and CD45R, bi-otinylated Abs to CD24 (M1/69) and CD43 (S7), and allophycocyanin-conjugated c-Kit Ab were purchased from PharMingen. The IL-7R cloneused in these studies was a gift from Richard R. Hardy (Institute for CancerResearch, Fox Chase Cancer Center, Philadelphia, PA).
Preparation of BM cells and isolation of hemopoieticstem/progenitor cell populations
The procedure for the preparation of BM cells for sorting has been previ-ously described (20). Briefly, BM cells were isolated from femurs and tibiaof donor mice, and the RBCs were lysed in an ammonium chloride potas-sium solution. The cells were incubated in a lineage cocktail containingoptimized concentrations of Abs to CD2, CD3, CD5, CD8, Mac-1, Gr-1,TER119, CD45R, and CD19. The CD45R Ab was not included in thelineage cocktail whenever CD45R expression was evaluated after lineagedepletion. Lineage depletion was conducted by two successive incubationsof the BM cells in sheep anti-rat Ig-coupled magnetic beads (Dynal, Oslo,Norway). The Linnegcells were stained with PE-Sca-1 and sorted using theFACSVantage (Becton Dickinson, San Jose, CA) set at enrichment modeand thresholding on PE emissions above background levels. Dead cellswere excluded from all analyses and sorts by gating on forward scatter andPI staining. The sorted LinnegSca-1poscells were pelleted and stained withallophycocyanin-c-Kit and FITC-Thy-1.1 and resorted into Thy-1.1lowc-Kitpos and Thy-1.1negc-Kitpos subsets. In experiments in which the Thy-1.1negc-Kitpos subset was further fractionated, the appropriate biotin-con-jugated Ab stain was added and visualized using Red613-SAv. All cellsorting steps were performed using the FACSVantage, and an aliquot of thesorted cell population was always taken for reanalysis.
Methylcellulose assays
The sorted cell populations were cultured in methylcellulose at a platingdensity of;100 cells/35-mm culture dish. Each milliliter of culture me-dium containeda-MEM (Life Technologies), 1.2% methylcellulose (Shi-netsu, Tokyo, Japan), 30% FCS (Life Technologies), 1% deionized BSA(Sigma, St. Louis, MO), and 0.1 mM 2-ME (Mallinckrodt, Chesterfield,MO) supplemented with the indicated cytokine combinations. Culturedishes were incubated at 37°C and infused with 5% CO2. The number ofcolonies was counted using an inverted microscope after 7 days of cultureto determine the cloning efficiency of each sorted cell population. Four tosix plates were scored for each group, and the results were expressed aspercentage of the total cells plated. In addition, individual colonies wereplucked between days 6 and12 of culture and analyzed for both cell surfacestaining and cell morphology. Two-thirds of the cells harvested from eachcolony were stained with Abs to CD45R and Gr-1 and analyzed by flowcytometry, while cytospins were prepared from the remaining cells. Cyto-spins were stained with May-Grunwald-Giemsa for morphologicalanalysis.
Liquid cultures
Liquid cultures of sorted cell populations were conducted usinga-MEM(Life Technologies) containing 10% FCS, 1 mM MEM sodium pyruvatesolution (Life Technologies), 10 mM HEPES (pH 7.3), 100 U/ml penicil-lin, 100 mg/ml streptomycin, 2 mM glutamine, and 0.1 mM 2-ME(Mallinckrodt) and supplemented with the indicated cytokine combina-
tions. Cells were either grown in bulk in 24-well plates or seeded at lim-iting dilution (one cell per well) in 96-well plates with or without stromalcell feeder layers as indicated. Culture plates were incubated at 37°C andinfused with 5% CO2. The 2018 stromal cell line (a gift from KateriMoore) was maintained at 31–33°C. Cells were prepared the day beforecoculture by seeding 10,000 cells/well in 24-well plates for bulk culturesand 1,000 cells/well in 96-well plates for the limited dilution (clonal) as-says. The presence of a single cell per well in 96-well plates was confirmedwhenever possible after overnight culture using an inverted microscope.Positive clones (wells) were scored by day 5 and harvested for analysisbetween days 6 and 14. For bulk cultures, representative wells were har-vested for analysis between days 5 and 14.
Intrathymic T cell development assay
Sublethally irradiated B6 (4- to 6-wk-old females) mice were anesthetizedand immobilized with rubber bands. The skin over the chest was incised toreveal the sternum, which was cut. The thymus was visualized within thethoracic cavity, and 3ml of fluid containing the sorted cell population ofinterest was directly injected into the thymic tissue using a Hamilton sy-ringe (Reno, NV). The chest was closed using stainless steel surgical clips.The cells to be transplanted were obtained from the B6-Thy-1.1-Ly-5.1double-congenic strain and were sorted directly into a microfuge tube con-taining a known amount of Hanks’ 10% FCS so that each 1ml of fluidcontained a known number of cells. Graded doses of cells were injectedinto groups of animals (10 animals/group) in the presence of an excess ofLinnegcells obtained from a second B6 congenic strain (B6-Ly-1.1), whichserved as a carrier and as an internal control to indicate successful injec-tions. Three or 4 wk later, the recipient B6 mice were sacrificed, and thy-mic tissue was isolated for analysis by flow cytometry to identify thymiclobes containing progeny cells derived from the injected populations. Suc-cessful intrathymic transfers were identified by the presence of Ly-1.1pos
cells, and positive thymic lobes were scored for the presence of Ly-5.1pos
cells. Limiting dilution statistics were applied to the resulting data to derivethe frequency of repopulating cells in the sorted population.
RT-PCR assay
Sorted cells were lysed using 500ml of TRIzol (Life Technologies) with 20mg of glycogen (Roche, Indianapolis, IN) added as a carrier. The TRIzolprotocol for RNA isolation prescribed by the manufacturer was followedusing half volumes. After isopropanol precipitation, the RNA pellet waswashed twice in 70% ethanol and resuspended in 8ml of diethylpyrocar-bonate-treated water. The RNA samples were incubated with 1ml of am-plification grade DNase I (Life Technologies) and 1ml of 103 DNase Ibuffer at room temperature for 15 min to eliminate any contaminatingDNA. The reaction was stopped with the addition of 1ml of 25 mM EDTAand heating at 72°C for 10 min. Water was added to bring the total volumeof each reaction to 20ml. Five to 10ml from each total RNA sample wasused for first-strand synthesis using random primers (Life Technologies)and Moloney murine leukemia virus reverse transcriptase (Life Technol-ogies) following the protocol provided by the manufacturer.
Semiquantitative PCR was used to compare the expression of genesbetween sorted cell populations. All primer sequences used in this studyhave been previously described. To equalize for cDNA input, each samplewas first amplified by PCR using GAPDH primers (21), and the amount ofinput cDNA was adjusted to provide equivalent signals. Subsequent PCRamplifications used the predetermined amount of cDNA with gene-specificprimers for sterile Ig heavy chain transcriptmo, Rag-2, E2A, Pax-5, andCD19 (7, 22, 23). PCR cycle parameters used for GAPDH,mo, and Rag-2were described by Li et al. (22), while those for E2A, Pax-5, and CD19were described by Bain et al. (23). Fifteen-microliter aliquots were with-drawn at cycles 24, 27, and 30 (GAPDH) or cycles 27, 30, and 33 (mo,Rag-2, E2A, Pax-5, and CD19) to assure that amplification was within thelinear range. The PCR products were separated by 1% agarose gel elec-trophoresis. Quantitation was performed using the MultiAnalyst program(Bio-Rad, Hercules, CA). Individual bands were measured and normalizedusing the GAPDH signal for each sample. Comparison of gene expressionbetween samples was achieved by comparing the normalized value foreach sample to the value obtained for CD45Rpos cells.
ResultsThe Thy-1.1neg cell population contains three separateprogenitor subsets
Linnegmouse BM cells were stained with Abs to Sca-1, c-Kit, andThy-1.1 and sorted to recover the Thy-1.1low and Thy-1.1neg cellpopulations (Fig. 1). The Thy-1.1neg subset comprised;0.05 6
0.01% (mean6 SD; n 5 8) of nucleated BM cells, a frequencyvery similar to that of the Thy-1.1low subset as previously reported(1, 2). Virtually all cells expressing low levels of Thy-1.1 werec-Kithigh, while the Thy-1.1neg population included cells express-ing both low and high levels of c-Kit. Cells lacking c-Kit expres-sion were not further characterized in these studies. Because ofconcerns regarding contamination of Thy-1.1neg cell preparationswith Thy-1.1low HSC, reanalysis of sorted populations was alwaysperformed as shown in Fig. 1, and Thy-1.1neg populations con-taining any discernable contamination with cells expressing Thy-1.1 at a level 5- to 10-fold above background levels were not usedfor functional studies. Although our previous transplant studiesdemonstrated an inhibitory influence of allophycocyanin-conju-gated c-Kit Abs on in vivo engraftment of Thy-1.1neg cells (19),direct comparisons of cloning efficiencies and lineage potentials ofThy-1.1neg cells isolated using biotin or allophycocyanin conju-gates of anti-c-Kit Abs showed no differences in our in vitrostudies.
Initial in vivo transplant studies demonstrated that the Thy-1.1neg cell population mediates rapid BM engraftment and con-
tributes to both lymphoid and myeloid lineages (19). To determinewhether the Thy-1.1neg cell subset consisted of multipotent pro-genitors or separate progenitor cells committed to either lymphoidor myeloid lineages, we conducted clonal progenitor cell assays.The sorted cells were cultured in methylcellulose or single-cellliquid cultures supplemented with different cytokine combinations,and individual colonies were analyzed after 6–12 days in cultureby both flow cytometry and cell morphology. These analysesshowed that the Thy-1.1neg cell subset contained three types ofprogenitors (Table I). Stimulation with a mixture of seven cyto-kines, as detailed in Table I, allowed differentiation of multiplehemopoietic lineages. Under these culture conditions, we observedcolonies consisting solely of myeloid lineage cells (macrophages,primitive granulocytes, and erythroid cells), colonies consistingsolely of lymphoid lineage cells (CD45RposGr-1neg), and mixedcolonies containing both lineages. Lymphoid lineage colonies rep-resented 33% of the total Thy-1.1neg colonies analyzed, whilethose consisting of myeloid lineage cells represented 54% (TableI). Mixed lineage colonies, containing CD45Rpos cells as well asmyeloid cells, were observed at a frequency of 13%. Similar re-sults were obtained in liquid cultures initiated from single cells,suggesting that the mixed lineage colonies were not the result ofsampling error in the methylcellulose assay. When Thy-1.1low
HSC were cultured under the same conditions, pure lymphoid col-onies were not observed, and very few mixed lineage coloniescontaining CD45Rpos cells were scored (4%, or 2 colonies of 45examined; Table I). Stimulation of Thy-1.1low HSC with lym-phoid-specific cytokines (STL, IL-7, and Flt3L) resulted in amarked decrease in cloning efficiency (88 vs 581 colonies/1000cells plated).
IL-3 has been reported to inhibit early B lymphoid differentia-tion (24, 25). Consistent with this observation, we observed thatomission of IL-3 from the cytokine cocktail resulted in a 40–50%decrease in the cloning efficiency of Thy-1.1neg cells. Analysis ofthe colonies that grew in the absence of IL-3 demonstrated a de-crease in the frequencies of pure myeloid and mixed colonies anda concomitant increase in the proportion, but not the absolute num-ber, of pure lymphoid colonies. Comparison between S7F6EGstimulation with or without IL-3 showed that 33% of 1486 44colonies were lymphoid in the presence of IL-3 (48.86 14 colo-nies/1000), whereas 60% of 816 29 colonies were lymphoid inthe absence of IL-3 (48.66 17 colonies/1000; Table I). Thy-1.1neg
cells grown in methylcellulose cultures supplemented with cyto-kines permissive only for lymphoid differentiation (S7F) exhibiteda 6-fold decrease in cloning efficiency compared with the morecomplex cytokine combination, and all colonies evaluated from
FIGURE 1. BM cells isolated from adult mice were lineage depletedand subsequently stained with mAbs to c-Kit, Sca-1, and Thy-1.1. Deadcells were excluded using propidium iodide staining and forward scattergating. Sca-1poscells were selected by gating (A) and analyzed for Thy-1.1expression (B). Thy-1.1negSca-1posc-Kitpos Linneg (Thy-1.1neg) and Thy-1.1low Sca-1posc-KitposLinneg (Thy-1.1low) cells were isolated by cell sort-ing as described inMaterials and Methods, and aliquots were taken forreanalysis (CandD).
Table I. Cloning efficiency and lineage potential of Thy-1.1low and Thy-1.1neg subsetsa
a Linneg BM cells were sorted to isolate Thy-1.1low Sca-1pos c-kitpos HSC (Thy-1.1low) or Thy-1.1neg Sca-1pos c-kitpos progenitor cells (Thy-1.1neg) as shown in Fig. 1.b Cultures were initiated in methylcellulose medium containing the indicated cytokines as described inMaterials and Methods.Cytokines are abbreviated as follows: S, STL;
7, IL-7; F, Flt3L; 3, IL-3; 6, IL-6; E, Epo; G, G-CSF.c Colonies were counted on days 7–9 of culture. Cultures initiated with Thy-1.1negcells stimulated with S7F contained 600 cells/plate; all other cultures were initiated with
120 cells/plate.d To evaluate lineage content, individual colonies were isolated from the methylcellulose medium and split for flow cytometric and histological analysis as described in
Materials and Methods.Of all colonies isolated from culture, 70% (Thy-1.1neg) to 90% (Thy-1.1low) contained enough cells for analysis. The percentage of the total and numberof colonies evaluated (n) are presented.
3044 DEFINITION OF EARLY B LYMPHOID-COMMITTED PROGENITORS
these cultures contained only cells with lymphoid morphology andsurface Ag expression. The presence of pure lymphoid coloniesafter stimulation of Thy-1.1neg cells with the complex cytokinemixture confirms that a committed lymphoid progenitor populationis contained within the Thy-1.1neg cell subset, but not in the Thy-1.1low HSC population. These cells grow in response to the cyto-kine combination of S7F and are insensitive to the inhibitory ef-fects of IL-3.
A number of early B lymphoid markers fractionate the Thy-1.1neg cell population into distinct subsets
To determine whether additional markers could potentially frac-tionate the Thy-1.1neg cell population into functionally distinctsubsets, we isolated Thy-1.1negcells and evaluated the expressionof a number of cell surface markers known to be expressed duringthe early stages of lymphoid development. Representative FACSplots of the staining analyses are shown in Fig. 2. As shown in Fig.2A, the AA4.1 mAb identifies a subset comprising 30–50% of theThy-1.1neg cells that expresses low levels of c-Kit. Expression ofIL-7R also separated the Thy-1.1neg population into two distinctclusters that correlated with c-Kit staining intensities. Cells thatwere IL-7Rpos invariably expressed low levels of c-Kit, while IL-7Rnegcells were both c-Kitlow and c-Kithigh (Fig. 2B). In contrast,CD62L/MEL14 expression was observed on both the c-Kitlow andthe c-Kithigh subsets of Thy-1.1negcells (Fig. 2C). Of the other cellsurface Ags tested, CD4 staining revealed only a minor subpopu-lation of positive cells (10% of Thy-1.1neg cells, equally distrib-uted between the c-Kitlow and c-Kithigh subsets, Fig. 2D), whichwas similar to the distribution of the Sca-2 Ag (data not shown).CD3 served as a negative control and was expressed by,1% ofThy-1.1neg cells (Fig. 2E). Since expression of AA4.1 and IL-7Rhave previously been associated with early lymphoid progenitors(4, 6, 7, 26), we focused on the c-Kitlow subset of cells for addi-tional phenotypic and functional analysis.
A hierarchy of progenitors in the B lymphocyte developmentalpathway has been described by Hardy and colleagues, who usedcell surface Ag expression to define specific developmental stages(4, 7, 27). To better place the Thy-1.1negc-Kitlow progenitor pop-ulation in the context of the previous studies, we evaluatedCD45R, CD24, and CD43 expression by these cells. Unlike theprimitive pro-B cell described by Hardy’s stage A, Thy-1.1negc-Kit low cells largely lack expression of CD45R and express highlevels of CD24 (Fig. 2,F andG). In common with Hardy’s stageA of development, Thy-1.1neg c-Kitlow cells express lower levelsof CD43 relative to Thy-1.1low stem cells (Fig. 2H). Hardy andcolleagues used expression of AA4.1 along with CD4, CD24, andCD43 to identify a CD45Rneg stage of B cell development pre-ceding the A stage (7), but the isolation protocol used in thosestudies failed to segregate these cells away from erythroid lineageprogenitors (4). Furthermore, functional assessment of B lineageprecursors in adult mouse bone marrow established that most earlyB lineage progenitors express CD24, as detected using either theM1/69 or 30-F1 Ab (9). To summarize, Thy-1.1negc-Kitlow cellsoverlap with LinnegTdT1 cells as defined by Tudor et al. (9), butdiffer from fraction A0 as defined by Hardy and colleagues (7) inthat most Thy-1.1negc-Kitlow cells lack CD4 and express high lev-els of CD24 (Fig. 2,D andG).
The lymphoid-committed progenitors in the Thy-1.1neg cellpopulation are predominantly c-Kitlow
To address the developmental potential of the individual subsets ofcells isolated in the Thy-1.1negpopulation, we used levels of c-Kit andAA4.1 expression as selection criteria to sort the Thy-1.1negpopula-tion into three subsets: c-KithighAA4.1neg, c-KitlowAA4.1neg, and
c-KitlowAA4.1pos(Fig. 2A). Methylcellulose and liquid cultures wereinitiated with each c-Kit/AA4.1 subset to ascertain their cloning effi-ciencies and myeloid and/or lymphoid differentiation potential. Thesecultures were supplemented with cytokine combinations selected toeither support proliferation and differentiation toward the lymphoidlineage (STL1IL-7 1 Flt3L, S7F), the myeloid lineage (IL-31 IL-61 Epo 1 G-CSF, 36EG), or both lymphoid and myeloid lineages(STL 1 IL-7 1 Flt3L 1 IL-3 1 IL-6 1 Epo1 G-CSF, S7F36EG).
FIGURE 2. BM cells from young (5- to 6-wk-old) adult mice were lin-eage depleted using an Ab cocktail as described inMaterials and Methods,substituting a CD19 monoclonal in place of CD45R. This population wassubsequently stained with mAbs to Sca-1, c-Kit, and Thy-1.1 and sorted forThy-1.1neg and Thy-1.1low cells as described in Fig. 1. Sorted cell popu-lations were then stained with mAbs to early lymphoid markers for sub-sequent analysis by FACS.A–E, Analyses of Thy-1.1neg CD45Rneg cells.F, The c-Kitlow subset defined in plotsA–E was selected by gating foranalysis of CD45R expression on c-Kitlow cells. G andH, Expression ofCD24 and CD43 by Thy-1.1low c-Kithigh HSC (u) compared with Thy-1.1neg c-Kitlow lymphoid progenitors (M). Plot I shows gating forCD45Rneg and CD45Rpos subsets of Thy-1.1neg c-Kitlow cells, which weredefined using a control stain of lymph node lymphocytes (u) to establisha sorting gate for negative and positive cells.
Evaluation of the cloning efficiencies of the subsets showed thatc-KithighAA4.1neg cells cloned at a frequency of 20% in S7F36EGand gave rise to myeloid or mixed colonies, but not pure lymphoidcolonies (Table II). Cloning efficiency dropped to 5% when the cellswere stimulated under lymphoid conditions. In contrast, the two c-Kit low populations resolved by AA4.1 staining showed equivalent col-ony growth in either lymphoid-specific or complex cytokine condi-tions, suggesting that the majority of clonogenic c-Kitlow cells arelymphoid committed (Table II). This interpretation was strengthenedby flow cytometric and cytospin analysis of individual colonies grownin S7F36EG. Of 99 separate c-KitlowAA4.1posand c-KitlowAA4.1neg
colonies analyzed, 88 (89%) contained only lymphoid lineage cellsdespite the presence of myeloid-promoting cytokines in the cultures.Mixed lineage colonies were most frequent in the c-KithighAA4.1neg
subset, where they represented 15% of the total colonies. These pro-genitors were also identified in both the AA4.1posand AA4.1negsub-sets of c-Kitlow cells, with their frequency being higher in theAA4.1neg subset (13% of total colonies, compared with 6% amongthe AA4.1poscells). Similar results were obtained in single-cell liquidcultures, suggesting that mixed lineage colonies were not due to in-advertant mixing of adjacent single-lineage colonies. Cytospin anal-ysis confirmed the presence of myeloid cells as detected by flow cy-tometry. A limited variety of myeloid cells were observed. Neutrophilmorphology was limited to very primitive-appearing cells with cres-cent nuclei. In contrast, mature-appearing macrophages were consis-tently observed in cytospin preparations. Purely myeloid coloniesarising from Thy-1.1negc-Kitlow cells were very rare (2% of total col-onies), and colonies containing erythroid lineage cells based on ben-zidine staining of hemoglobin were never observed in the myeloid ormixed colonies derived from Thy-1.1negc-Kitlow cells. These resultsconfirm previous observations that pro-B cells are characterized by alow level of c-Kit expression (9, 28), and extend these findings byshowing that these early pro-B cells can be completely separated fromprogenitors for nonlymphoid hemopoietic lineages using the Thy-1.1neg selection protocol.
Mixed lineage colonies generated by Thy-1.1negc-Kithigh cellsmaintained both macrophage and lymphoid progeny over the lifeof the cultures. In contrast, the mixed lineage colonies detected incultures initiated with Thy-1.1negc-Kitlow cells (either AA4.1posorAA4.1neg) were transient, as they could only be detected on day 6or 7 of culture. A comparison of the growth kinetics of mixedlineage colonies in the c-KitlowAA4.1pos and c-KitlowAA4.1neg
cell subsets is shown in Fig. 3A. The c-KitlowAA4.1negsubset didnot produce visible colonies until day 7 in culture. At this time, anequivalent number of pure lymphoid and mixed colonies was ob-
served. On subsequent days, only pure lymphoid colonies weredetected. A similar pattern was observed when c-KitlowAA4.1pos
cells were cultured, except that colonies appeared 1 day earlier,and pure lymphoid colonies always outnumbered mixed colonies.Thus, the mixed lineage potential of the Thy-1.1neg c-Kitlow sub-sets of cells was limited, and detection of the myeloid progeny wastransient.
Table II. Cloning efficiency and lineage potential of Thy-1.1neg subsets defined by c-Kit and AA4.1a
a LinnegBM cells were sorted to isolate Thy-1.1negSca-1posc-kitposprogenitor cells as shown in Fig. 1. This population was further fractionated into the indicated populationsas shown in Fig. 2Aand 3A.
b Cultures were initiated in methylcellulose medium containing the indicated cytokines as described inMaterials and Methods.Cytokines are abbreviated as follows: S, STL;7, IL-7; F, Flt3L; 3, IL-3; 6, IL-6; E, Epo; G, G-CSF.
c Colonies numbers were evaluated on day 8 of culture.d Lineage content was evaluated as in Table I on days 6–12 of culture.
FIGURE 3. A, Observed kinetics of lineage outcomes of coloniesplucked from sorted Thy-1.1neg c-Kitlow AA4.1 subsets after 6–9 days ofculture in methylcellulose supplemented with S7F36EG.M andf, Colo-nies from Thy-1.1neg c-KitlowAA4.1neg subset;E and F, colonies fromThy-1.1negc-KitlowAA4.1possubset.M andE, Lymphoid lineage colonies;f andF, mixed lineage colonies.B, Observed cell surface staining patternof sorted Thy-1.1neg c-KitlowAA4.1neg cells after 5–8 days of culture inmedium supplemented with S7F36EG. Most of the Thy-1.1neg c-Kitlow
AA4.1neg cells became AA4.1pos and CD45Rpos by day 5 and retainedexpression of c-Kit. By day 8, virtually all cells were CD45RposGr-1neg.
3046 DEFINITION OF EARLY B LYMPHOID-COMMITTED PROGENITORS
The kinetics of colony evolution shown in Fig. 3A suggest thatthe c-KitlowAA4.1negand the c-KitlowAA4.1possubsets form a lin-eage, with c-KitlowAA4.1neg cells differentiating to generate themore lineage-restricted c-KitlowAA4.1poscells. To address this hy-pothesis, Thy-1.1neg c-KitlowAA4.1neg cells were grown as repli-cate bulk cultures in the presence of S7F36EG in 24-well plates.Representative wells were harvested over a period of 8 days, andthe cells were stained with mAbs to AA4.1, CD45R, c-Kit, andGr-1. As shown in Fig. 3B, most of the cells isolated in the Thy-1.1neg c-KitlowAA4.1neg population become AA4.1pos andCD45Rpos after 5 days in culture. These cultures included rarecells expressing the myeloid lineage marker (Gr-1) on day 5 (datanot shown) and day 6, but became predominantly CD45Rpos byday 8. These results, although not at a clonal level, recapitulate thekinetics of CD45R and Gr-1 cell surface expression observed inclonal colonies derived from the c-KitlowAA4.1negcells (Fig. 3A).This experiment also provides further evidence that thec-KitlowAA4.1negand the c-KitlowAA4.1possubsets form a lineage.
It is interesting to note that virtually all cells in these culturescoexpressed AA4.1, CD45R and c-Kit at all time points (Fig. 3Band data not shown). In contrast, parallel cultures initiated withThy-1.1negc-Kithigh cells included a heterogeneous pattern of c-Kitexpression and few CD45Rposc-Kitposcells (data not shown). Thisresult underscores the multilineage potential of the Thy-1.1neg c-Kithigh subset in contrast to the predominantly lymphoid-commit-ted Thy-1.1neg c-Kitlow subsets.
As shown in Fig. 2F, the majority of cells defined by the phe-notype Thy-1.1neg c-KitlowAA4.1pos do not express CD45R, andour clonal assays indicate that this cell subset consists almost ex-clusively of lymphoid-committed progenitors (Table II and Fig.3A). However, the efficiency of magnetic bead depletion may notbe absolute, particularly for cells expressing low levels of the tar-get Ag. In addition, the cloning efficiency of Thy-1.1neg
c-KitlowAA4.1poscells (;5%) was equivalent to the low frequencyof CD45Rpos cells contained in the Thy-1.1neg c-KitlowAA4.1pos
subset (Table II). To ascertain that lymphoid lineage potential isindeed contained within the CD45Rneg fraction of thec-KitlowAA4.1pos cells, we replaced anti-CD45R with anti-CD19for magnetic lineage depletion and sorted these Linneg cells forCD45Rneg and CD45Rpos fractions within the Thy-1.1neg
c-KitlowAA4.1pos cell subset (Fig. 2I). The cloning efficiencies ofboth CD45R fractions were comparable (5.8% for CD45Rneg,4.9% for CD45Rpos), and 96% of colonies assayed (85 of 88) fromthe CD45Rnegc-Kitlow AA4.1poscell subset were lymphoid in lin-eage (data not shown). This result suggests that the Thy-1.1neg
c-Kitlow subset represents a developmental stage at which surfaceexpression of CD45R is just beginning to be induced.
To investigate the influence of stromal cell monolayers on thecloning efficiency of the Thy-1.1neg c-Kitlow cell subsets, we es-tablished liquid cultures in the presence or the absence of clonedstromal cell lines. Single cells isolated from the AA4.1neg andAA4.1pos subsets of Thy-1.1neg c-Kitlow cells by automatic celldeposition FACS sorting were seeded into microtiter wells con-taining cytokines alone (S7F36EG or S7F) or in the presence ofS7F plus three different stromal lines reported to support B lym-phoid development (Fig. 4A). Any of three different bone marrowstromal cell lines (S17 (29), AC6–2.1 (30), and 2018 (31)) in thepresence of exogenous S7F enhanced the cloning efficiency ofAA4.1poscells by about 2-fold, while the AA4.1negsubset culturedwith S7F and the 2018 stromal line exhibited only a slight en-hancement in cloning efficiency compared with S7F stimulation inthe absence of stromal cells (Fig. 4A). Individual clones were har-vested and evaluated for expression of surface Ags by flow cy-tometry. On days 11 and 12 of culture, single cells cocultured with
2018 in the presence of S7F had expanded to colonies rangingfrom 104 to 3 3 104 cells and expressed CD45R, CD24, and BP1uniformly and CD43 at variable levels. Surface IgM (sIgM) ex-pression was observed on very few cells (data not shown). Bulkcultures established from 53 102 AA4.1neg or AA4.1pos cells inS7F with or without 2018 cells expanded to 350- to 900-fold (60–70% viable) in 7 days and 1200-fold (30% viable) in 14 days,demonstrating the extensive proliferative potential of these cells.After 14 days in culture, sIgM expression could be detected on asmall subset of cells growing either in cytokines alone or on 2018cells in the presence of S7F (Fig. 4B). However, only about 10%of the cells in the sIgMneg population expressed cytoplasmic IgMon day 11, supporting the interpretation that most of the expandingcells in the cultures represent pro-B cells. This result is also con-sistent with recent studies suggesting that high levels of IL-7 leadto expansion of cIgMneg pro-B cells, and that selective regulationof IL-7 dose-response thresholds leads to preferential outgrowth ofpre-B cells expressing productively rearranged IgM in associationwith the surrogate light chains and signaling components of thepre-B cell receptor (32). These results demonstrate high cloning
FIGURE 4. A, Cloning efficiencies of sorted c-Kit AA4.1 subsets of theThy-1.1neg cell population grown in medium supplemented with differentcombinations of cytokines (S7F36EG and S7F) or cocultured with differentstromal cell lines (AC6, S17, and 2018) supplemented with S7F. AC6 andS17 cultures were performed once, while all others represent mean resultsfrom three separate experiments6 SEM. In the case of AA4.1negcells, theerror in measurement of cloning efficiency was too small to plot.B, Flowcytometric analysis of bulk cultures of 500 AA4.11 cells grown in mediumsupplemented with S7F (upper panels) or cocultured with 2018 stromalcells supplemented with S7F (lower panels) for 7 or 14 days. Among viablecells, 86% of cells derived from both cytokine-driven and stromal cellcocultures express CD45R on day 7, and this increased to 99% by day 14.The percentage of CD45Rposcells expressing sIgM is indicated. On day 11,10% of cells cultured with 2018 and S7F were positive for cytoplasmicIgM (data not shown).
efficiency and proliferative potential of the Thy-1.1negc-Kitlow cellsubsets.
The T cell potential of the Thy-1.1neg cell population ispredominantly found in the c-KitlowAA4.1neg subset
To assess whether the lymphoid-committed Thy-1.1neg subsetshave committed to the B lymphocyte lineage or, alternatively, areanalogous to the previously reported common lymphoid progenitor(6), T cell progenitor cell assays were performed. Limiting dilutionanalysis was performed by injecting graded numbers of cells(0–85 c-KitlowAA4.1neg cells and 0–192 c-KitlowAA4.1pos cells)into thymic lobes of sublethally irradiated mice and scoring thelobes as positive or negative for T cell development 15–25 dayslater. To control for the efficiency of thymic injections, we coin-jected the sorted cell subset of interest (isolated from B6 congenicmice carrying the Thy-1.1, Ly-1.2, and Ly-5.1 alleles) along witha saturating dose of Linneg cells derived from a second B6-con-genic strain (Thy-1.2, Ly-1.1, Ly-5.2) into B6 mice (Thy-1.2, Ly-1.2, Ly-5.2). Successful intrathymic transfers were identified bythe presence of Ly-1.1poscells, and these thymic lobes were scoredfor the presence of Ly-5.1pos cells. This analysis showed that thefrequency of T cell progenitors within the c-KitlowAA4.1neg andc-KitlowAA4.1poscell subsets was 1 in 160 and,1 in 620, respec-tively (Fig. 5). This result supports the interpretation that expres-sion of AA4.1 coincides with B lineage commitment, since a fre-quency of 1 in 620 could be explained by contamination withpopulations of cells outside our sorting gates. The low frequencyof T cell progenitors in the c-KitlowAA4.1neg population suggeststhat common lymphoid progenitors can account for only a subsetof these cells.
Expression of B lineage-associated genes in the Thy-1.1neg
c-Kitlow lymphoid-committed progenitor subsets
The results of the in vitro clonal assays and the intrathymic injec-tions suggest that the majority of c-KitlowAA4.1pos cells are com-mitted to the B lymphoid lineage at a developmental stage coin-cident with up-regulation of CD45R expression. Commitment tothe B lymphoid lineage occurs before Ig heavy chain rearrange-ment (4) and is associated with transcriptional activation of thislocus, resulting in the expression of germlinem transcripts (mo) (7)as well as the expression of a number of genes that are required for
Ig gene rearrangements (recombinase-activating gene (Rag)-1 andRag-2) (22). Transcription ofmo has been shown to reflect theaccessibility and competence of them region for Ig gene rearrange-ments (7, 33). In addition, recent studies have established the im-portance of a number of transcription factors in lymphoid devel-opment (34). Three of these proteins, products of thePax5,E2A,andEBF genes, are essential in both B cell lineage commitmentand B cell development (23, 35–37). To better establish the stageof development represented by the two AA4.1 subsets of Thy-1.1neg c-Kitlow cells, we performed gene expression analysis bysemiquantitative RT-PCR. We determined the expression ofmo,Rag-2, E2A, Pax5, and CD19 in the various Thy-1.1neg cell pop-ulations compared with CD45Rpos BM cells (Fig. 6). Rag-2 ex-pression was detected at equivalent levels in all samples. In con-trast,mo transcripts were highly expressed by Thy-1.1neg c-Kitlow
cells and by the two AA4.1 subsets relative to the total populationof CD45R1 cells. Since the target sequence for the 59oligonucle-otide used to prime this amplification is deleted upon IgH D-Jrearrangement (7, 33), this result demonstrates that both AA4.1subsets include cells with at least one transcriptionally active alleleof the IgH gene in germline configuration. Fig. 6 also providesfurther evidence supporting the conclusion that AA4.1negcells arethe precursors of AA4.1poscells. Although each population equallyexpressed the E2A gene, both Pax5 and CD19 were up-regulatedin AA4.1pos cells relative to AA4.1neg cells. This observation isconsistent with transfection experiments that indicate induction ofPax5 by E2A and of CD19 by Pax5 (38–40). Thus, the Thy1.1neg
c-KitlowAA4.1neg stage of development corresponds to the aber-rant B cell progenitors in Pax5-deficient mice that can be lineageredirected using IL-7 and other cytokines (36). Together with ourfunctional data, these molecular results strongly support the con-clusion that the Thy-1.1neg c-Kitlow subsets include cells at theearliest stages of B cell development in the mouse.
DiscussionTransplant studies have previously shown that the Thy-1.1negpop-ulation is enriched for committed lymphoid progenitors that canreadily engraft in the recipient BM (19). Here we demonstrate thatthe Thy-1.1neg cell population contains three progenitor popula-tions, including separate committed progenitors for both lymphoid
FIGURE 5. Analysis of T cell progenitor frequency in the Thy-1.1neg
c-KitlowAA4.1neg(f) and AA4.1pos(M) populations. Cell populations iso-lated from Ly-5.1 donor mice were injected into the thymic lobes of sub-lethally irradiated B6 animals along with Linneg BM derived from Ly-1.1congenic animals, as described inMaterials and Methods. After 12–20days, animals were sacrificed, and thymic lobes were analyzed for thepresence of Ly-1.11 cells to indicate successful injections. Lobes werescored as positive when a clone of Ly-5.11 cells.2% of the total thymicpopulation was detected. By limiting dilution analysis, the AA4.1negsubsetcontained 1 thymic progenitor/160 cells (r5 0.991), while the AA4.1pos
population contained,1 thymic progenitor/620 cells.
FIGURE 6. RT-PCR analysis of early B lineage genes expressed by thesubsets of cells described in these studies. Samples were adjusted based onGAPDH signal and amplified as described inMaterials and Methods. Ali-quots of PCR products were withdrawn at cycles 24, 27, and 30 (GAPDH)or cycles 27, 30, and 33 (mo, Rag-2, E2A, Pax-5, and CD19) and separatedby 1% agarose gel electrophoresis for detection by ethidium bromide stain-ing. The bar graphs depict expression levels relative to that detected inCD45Rpos BM cells, as assessed by densitometry scanning of the originalgel.A, Thy-1.1neg; B, Thy-1.1negc-Kitlow; C, Thy-1.1negc-KitlowAA4.1neg;D, Thy-1.1negc-KitlowAA4.1pos; E, CD45RposBM cells. Thus,B representsthe sum ofC andD.
3048 DEFINITION OF EARLY B LYMPHOID-COMMITTED PROGENITORS
and myeloid lineages as well as a mixed lineage progenitor. Com-pared with the Thy-1.1low HSC population, Thy-1.1neg cells havea more restricted pattern of myeloid lineage potential because mostcolonies contained macrophages, while mature neutrophils withsegmented nuclei were rare. Moreover, the Thy-1.1neg cell popu-lation could be further segregated based on the level of c-Kit andAA4.1 expression. Using these Ags, we delineated two function-ally distinct populations: c-KithighAA4.1neg cells, which are pre-dominantly committed myeloid progenitors, and c-KitlowAA4.1neg/pos
cells, which are predominantly lymphoid-committed progenitors (Ta-ble II). Thy-1.1neg c-Kitlow cells largely share the CD45Rneg pheno-type of the fraction A0 cells isolated by Allman et al. (4). However, incontrast to fraction A0, Thy-1.1negc-Kitlow cells are CD24pos, CD4neg
and do not include erythroid lineage progenitors. Tudor and col-leagues have shown that the majority of cells comprising fraction Ahave a low cloning efficiency and thus probably do not represent amajor intermediate in B cell development (9). The similarity in phe-notype, frequency, cloning efficiency, and differentiation potential ofThy-1.1neg c-Kitlow cells compared with the Linneg c-KitlowIL-7RaposFlk-2posCD34negcells described by Tudor and colleagues sug-gests that the two isolation protocols define a similar population ofcells. However, the selection protocol described in the present studiesmay provide a better separation of lymphoid and myeloid progenitors,since up to 30% of clones grown on S17 stromal cells by Tudor andcolleagues were of myeloid lineage. It is likely that the selectionagainst Thy-1.1 expression in the present studies accounts for theincreased resolution of B lymphoid progenitors from multipotent stemcells and myeloid progenitors.
Recently, a common lymphoid progenitor with the phenotypeThy-1.1negSca-1low c-KitlowLinnegIL7Rpos has been reported (6).This cell population possesses rapid and prominent lymphoid-re-stricted potential with limited or no self-renewal activity. Sincemost of the cells in the Thy-1.1neg c-Kitlow subset express IL-7R(Fig. 2B), this population of cells overlaps almost entirely with thecommon lymphoid progenitor of Kondo and colleagues. In con-trast to the findings of that group, we demonstrate that T lineagepotential is low among Thy-1.1neg c-KitlowAA4.1neg cells and isabsent from the Thy-1.1neg c-KitlowAA4.1pos subset (Fig. 5). Oneobvious difference between the two sets of experiments is theshort-term culture in S7F used by Kondo and colleagues to provethat the clonal progeny of single cells could differentiate into boththe T and B lineages. A similar observation was reported by Ja-cobsen and colleagues, who showed that c-KithighSca-1posLinneg
HSC could be cultured for up to 2 weeks in IL-7 and Flt3L and thatthe cultures, when transplanted i.v., reconstituted T and B, but notmyeloid, lineages (41). Older studies by Phillips and colleaguesreported similar findings using long term bone marrow culturestransplanted into immunodeficient recipient mice (42). These re-sults are also consistent with studies using animals mutated at thePax5 locus, since pro-B cells from these animals could be culturedlong term in IL-7 while retaining T lineage potential (43). Takenwith the results shown in Fig. 5, these observations indicate thatcommon lymphoid progenitors will default to the B lineage unlessspecific stimulation with cytokines precedes thymic engraftment.This is consistent with an inductive mechanism for the T lineage,rather than a permissive or stochastic developmental pathway.
Kondo and colleagues proposed two pathways of developmentfor conventional B lymphocytes. One of these branches early fromthe HSCs, has the potential to develop in the T or B lineage but hasno myeloid potential, and expresses IL-7R. The other pathwaybranches later, has potential to develop as B lineage and a limitedarray of myeloid cells, and lacks the IL-7R. Although the Thy-1.1neg cell populations reported here as well as that reported byKondo and colleagues are capable of rapid BM reconstitution, giv-
ing rise to predominantly lymphoid lineage cells, the Thy-1.1neg
c-Kithigh cell subset as isolated in these studies retains some my-eloid differentiation potential. Thus, we suggest that the mixedlineage progenitor described within the Thy-1.1neg c-Kithigh cellsubset (Table II) may represent the second branch of B cell pro-genitors postulated by Kondo and colleagues, while the Thy-1.1neg
c-KitlowAA4.1neg cell population described here may include thecommon lymphoid progenitors described by Kondo and colleagues. Itis not clear to what extent each of these mutually exclusive interme-diates contributes to the Thy-1.1negc-KitlowAA4.1poscell subset andsubsequent B lineage development (Fig. 7).
Our analysis of T cell potential in the Thy-1.1negcell populationshows that these cells can differentiate in the thymus after intra-thymic injection (Fig. 5). Most of the T cell potential is containedin the Thy-1.1neg c-KitlowAA4.1neg cell subset. However, the fre-quency of cells in this subset that engraft intrathymically is 32-foldlower than the frequency of T cell progenitors found when Thy-1.1low HSC are injected (2). This suggests that Thy-1.1negc-Kitlow
cells may predominantly be committed to the B cell lineage andthat T cell progenitors may largely lie outside this subset of cells.Alternatively, the low frequency of T cell clones observed relativeto Thy-1.1low HSC may represent a difference in proliferative ex-pansion before the onset of TCR rearrangements. T cell clonesderived from progenitors at later stages of development would ap-pear earlier and would be smaller and may not be observed if theyfail positive or negative selection before becoming large enough tobe detected in this assay. Most studies indicate that the rate ofemigration of progenitor cells into the thymus is quite low (45).Several lines of evidence suggest that it is a committed progenitorrather than a pluripotent HSC that seeds the thymus to initiate Tcell development (46–48). We therefore expect to find a BM pro-genitor with sufficient proliferative activity to maintain productionof T lineage cells in the thymus in the absence of large scalereplenishment from marrow-derived sources. The sizes of thymic
FIGURE 7. The developmental relationships between the lymphoidprogenitors described in this study and other identified intermediates inearly lymphoid development. The HSC has been previously characterizedas Thy-1.1low and negative for a variety of lineage Ags, including CD45R,but positive for c-Kit, Sca-1, CD24, and CD43. Although stem cells in yolksac and fetal liver express AA4.1, adult BM stem cells do not (44). Wehave not established to what extent the pro-B cells described in this studyare progeny of the common lymphoid progenitors (CLP) described byKondo et al. (6) or a bipotent B/macrophage progenitor as described bynumerous investigators. The T and B lineage potential of the pro-B tran-sitional stage was derived from the frequencies of each activity (1 T pro-genitor/160 cells vs 1 B progenitor/20 cells). The CLP may be representedby the 10% of T lineage progenitors in the pro-B transitional stage de-scribed here, since the phenotypes of these two populations largely overlap.
clones should reflect the amount of proliferation they can undergobefore rearrangement of TCR genes and subsequent selection. Ad-ditional experiments to localize T cell progenitors immediatelyoutside the HSC compartment will be addressed in the immediatefuture.
A number of investigators have characterized early B cell de-velopment by isolating and characterizing early progenitors basedon CD45R expression. The earliest of these progenitors is thoughtto lack CD24 expression (7, 49, 50) and to up-regulate this markeras differentiation proceeds. The Thy-1.1negc-Kitlow cell populationisolated in these studies has strong hallmarks of being a transi-tional cell between the HSC and the progenitors for lymphoid lin-eages. Like the HSC, the Thy-1.1neg c-Kitlow progenitor isCD45Rneg and expresses CD24, with the CD24 expression beingslightly higher compared with HSC (Fig. 2G). In addition, Thy-1.1neg c-Kitlow cells will proliferate in response to the lymphoid-selective cytokine combination of S7F, exhibiting a cloning effi-ciency of 4–8% under these conditions (Table II and Fig. 4A). Weobserved that cytokine-driven bulk cultures of Thy-1.1neg
c-KitlowAA4.1pos cells, in the presence or the absence of stromalcells, supported lymphoid differentiation up to theCD45RposIgMneg B cell stage, but were inefficient at supportingfurther differentiation (Fig. 4B). Modulation of IL-7 concentrationsin these cultures may be necessary to select the rare cells thatsuccessfully complete Ig gene rearrangement (32, 51).
Recent studies have established the importance of a number oftranscription factors in lymphoid development (52). Three of theseproteins, products of thePax5,E2A, andEBF genes, are specifi-cally associated with commitment to the B cell lineage (53–55).Targeted mutation ofE2A or EBF results in a block in B celldevelopment at a stage before Ig gene rearrangements, whilePax5-deficient mice initiate D-JH rearrangement but arrest beforeV-DJH rearrangements (37). Interestingly, Pax-5 deficiency leadsto the growth of pro-B cells at a stage of development that allowslineage reprogramming even after extensive culture in the presenceof IL-7 (36). Thus, cultured Pax5-deficient cells that are pheno-typically identical with native pro-B cells and have rearrangedtheir Ig D-JH gene segments can subsequently be induced to dif-ferentiate as T lineage lymphocytes or even as macrophages, neu-trophils, osteoclasts, dendritic cells, and NK cells (43). This lin-eage reprogramming requires withdrawal of IL-7 and substitutionof other lineage-inducing cytokines or microenvironments, sug-gestive of a deterministic regulation of differentiation at this stageof development. Our results suggest that Thy-1.1neg c-Kitlow
AA4.1negcells represent a normal counterpart to the pseudo-pro-Bcell population found in Pax5-deficient animals, and that the de-velopmental stage at which Pax5 expression is normally up-regu-lated during B cell development is associated with up-regulation ofAA4.1 (Fig. 6). Enrichment of lymphoid progenitors within theThy-1.1neg cell population using c-Kit and AA4.1 as additionalphenotypic markers is further supported by the marked increase inone of the earliest markers of B lineage commitment,mo, that wasobserved in the lymphoid progenitor subsets compared with eitherthe whole Thy-1.1neg cell population or the more matureCD45Rpos cells (Fig. 6). Furthermore, the onset of Pax-5 andCD19 transcription at the AA4.1pos stage of development corre-sponds to very recent activation of these genes, since the cell pop-ulation does not yet express surface CD19 protein (data notshown).
The role of cytokines in supporting cell survival and promotingcell proliferation of HSCs has been the focus of several studiesover the past few years (56). In addition, there is considerabledebate about whether cytokines can directly influence lineagecommitment decisions of progenitor cells (57, 58). Our data sug-
gest that specific cytokine combinations can predictably influencethe lineage outcomes of sorted Thy-1.1neg cells. The absence ofIL-3 in the cytokine combination led to a marked decrease in my-eloid lineage colonies, while the combination of S7F specificallysupported lymphoid lineage colonies (Table I). In both cases, how-ever, there was a decrease in cloning efficiency and no markedincrease in the number of lymphoid colonies analyzed comparedwith the complex cytokine combination (S7F36EG). These obser-vations led to the identification of separate committed progenitorcells for lymphoid and myeloid lineages contained within the Thy-1.1neg cell population (Table II). The results also imply that thecombination of S7F selectively supports the proliferation and dif-ferentiation toward the lymphoid lineage of lymphoid-committedprogenitor cells, but is not sufficient to influence lineage commit-ment decisions of mixed lineage progenitor cells. Similar resultshave been reported for the role of IL-7 in lineage commitmentdecisions of bipotent lymphoid-myeloid progenitor cells frommouse fetal liver (59). The failure of apoptosis-inhibiting mole-cules to support continuing differentiation of B lineage cellsstrongly supports a selective role for extrinsic signals in lymphoiddevelopment (60).
Finally, it should be noted that erythroid and megakaryocytelineage cells were rarely observed in the methylcellulose culturesof Thy-1.1negcells and failed to recover in transplant studies. Thisobservation that the Thy-1.1neg cell subset gives rise to lympho-cytes, macrophages, and granulocytes, but not erythrocytes andmegakaryocytes, provides evidence for the early separation of themyeloid-lymphoid progenitors from the erythroid-megakaryocyteprogenitors. Several other laboratories have reported similar find-ings recently (3, 61–63). This is in contrast to the conventionalview of the hierarchy of hemopoietic differentiation, where thelymphoid progenitors are usually depicted as separate from pro-genitors of the myeloid, megakaryocyte, and erythroid lineagesearly in the differentiation process. Cell isolation studies such asthose reported here will be instrumental in defining the lineagerelationships at early stages of hemopoietic development that untilthis point have remained elusive.
References1. Morrison, S. J., N. Uchida, and I. L. Weissman. 1995. The biology of hemato-
poietic stem cells.Annu. Rev. Cell. Dev. Biol. 11:35.2. Spangrude, G. J., S. Heimfeld, and I. L. Weissman. 1988. Purification and char-
acterization of mouse hematopoietic stem cells.Science 241:58.3. Akashi, K., D. Traver, T. Miyamoto, and I. L. Weissman. 2000. A clonogenic
common myeloid progenitor that gives rise to all myeloid lineages.Nature 404:193.
4. Allman, D., J. Li, and R. R. Hardy. 1999. Commitment to the B lymphoid lineageoccurs before DH-JH recombination.J. Exp. Med. 189:735.
5. Payne, K. J., K. L. Medina, and P. W. Kincade. 1999. Loss of c-kit accompaniesB-lineage commitment and acquisition of CD45R by most murine B-lymphocyteprecursors.Blood 94:713.
6. Kondo, M., I. L. Weissman, and K. Akashi. 1997. Identification of clonogeniccommon lymphoid progenitors in mouse bone marrow.Cell 91:661.
7. Li, Y. S., R. Wasserman, K. Hayakawa, and R. R. Hardy. 1996. Identification ofthe earliest B lineage stage in mouse bone marrow.Immunity 5:527.
8. Rolink, A., E. ten Boekel, F. Melchers, D. T. Fearon, I. Krop, and J. Andersson.1996. A subpopulation of B2201 cells in murine bone marrow does not expressCD19 and contains natural killer cell progenitors.J. Exp. Med. 183:187.
9. Tudor, K. S., K. J. Payne, Y. Yamashita, and P. W. Kincade. 2000. Functionalassessment of precursors from murine bone marrow suggests a sequence of earlyB lineage differentiation events.Immunity 12:335.
10. Lu, L., and D. G. Osmond. 1997. Apoptosis during B lymphopoiesis in mousebone marrow.J. Immunol. 158:5136.
11. Spangrude, G. J., and G. R. Johnson. 1990. Resting and activated subsets ofmouse multipotent hematopoietic stem cells.Proc. Natl. Acad. Sci. USA 87:7433.
12. Uchida, N., L. Jerabek, and I. L. Weissman. 1996. Searching for hematopoieticstem cells. II. The heterogeneity of Thy-1.1lowLin2/lowSca-11 mouse hemato-poietic stem cells separated by counterflow centrifugal elutriation.Exp. Hematol.24:649.
13. Li, C. L., and G. R. Johnson. 1992. Rhodamine123 reveals heterogeneity withinmurine Lin2, Sca-11 hemopoietic stem cells.J. Exp. Med. 175:1443.
3050 DEFINITION OF EARLY B LYMPHOID-COMMITTED PROGENITORS
14. Fleming, W. H., E. J. Alpern, N. Uchida, K. Ikuta, G. J. Spangrude, andI. L. Weissman. 1993. Functional heterogeneity is associated with the cell cyclestatus of murine hematopoietic stem cells.J. Cell Biol. 122:897.
15. Nibley, W. E., and G. J. Spangrude. 1998. Primitive stem cells alone mediaterapid marrow recovery and multilineage engraftment after transplantation.BoneMarrow Transplant. 21:345.
16. Osawa, M., K. Nakamura, N. Nishi, N. Takahasi, Y. Tokuomoto, H. Inoue, andH. Nakauchi. 1996. In vivo self-renewal of c-Kit1 Sca-11 Linlow/2 hemopoieticstem cells.J. Immunol. 156:3207.
17. Okada, S., H. Nakauchi, K. Nagayoshi, S. Nishikawa, Y. Miura, and T. Suda.1991. Enrichment and characterization of murine hematopoietic stem cells thatexpress c-kit molecule.Blood 78:1706.
18. Orlic, D., R. Fischer, S. Nishikawa, A. W. Nienhuis, and D. M. Bodine. 1993.Purification and characterization of heterogeneous pluripotent hematopoieticstem cell populations expressing high levels of c-kit receptor.Blood 82:762.
19. Searles, A. E., S. J. Pohlmann, L. J. Pierce, S. S. Perry, W. B. Slayton,M. P. Mojica, and G. J. Spangrude. 2000. Rapid, B lymphoid-restricted engraft-ment mediated by a primitive bone marrow subpopulation.J. Immunol. 165:67.
20. Spangrude, G. J., D. M. Brooks, and D. B. Tumas. 1995. Long-term repopulationof irradiated mice with limiting numbers of purified hematopoietic stem cells: invivo expansion of stem cell phenotype but not function.Blood 85:1006.
21. Burn, T. C., M. S. Petrovick, S. Hohaus, B. J. Rollins, and D. G. Tenen. 1994.Monocyte chemoattractant protein-1 gene is expressed in activated neutrophilsand retinoic acid-induced human myeloid cell lines.Blood 84:2776.
22. Li, Y. S., K. Hayakawa, and R. R. Hardy. 1993. The regulated expression of Blineage associated genes during B cell differentiation in bone marrow and fetalliver. J. Exp. Med. 178:951.
23. Bain, G., E. C. Maandag, D. J. Izon, D. Amsen, A. M. Kruisbeek,B. C. Weintraub, I. Krop, M. S. Schlissel, A. J. Feeney, and M. van Roon. 1994.E2A proteins are required for proper B cell development and initiation of im-munoglobulin gene rearrangements.Cell 79:885.
24. Hirayama, F., S. C. Clark, and M. Ogawa. 1994. Negative regulation of early Blymphopoiesis by interleukin 3 and interleukin 1a. Proc. Natl. Acad. Sci. USA91:469.
25. Ball, T. C., F. Hirayama, and M. Ogawa. 1996. Modulation of early B lympho-poiesis by interleukin-3.Exp. Hematol. 24:1225.
26. McKearn, J. P., C. Baum, and J. M. Davie. 1984. Cell surface antigens expressedby subsets of pre-B cells and B cells.J. Immunol. 132:332.
27. Hayakawa, K., Y. S. Li, R. Wasserman, S. Sauder, S. Shinton, and R. R. Hardy.1997. B lymphocyte developmental lineages.Ann. NY Acad. Sci. 815:15.
28. Rolink, A., D. Haasner, S. Nishikawa, and F. Melchers. 1993. Changes in fre-quencies of clonable pre B cells during life in different lymphoid organs of mice.Blood 81:2290.
29. Henderson, A. J., A. Johnson, and K. Dorshkind. 1990. Functional characteriza-tion of two stromal cell lines that support B lymphopoiesis.J. Immunol. 145:423.
30. Whitlock, C. A., G. F. Tidmarsh, C. Muller-Sieburg, and I. L. Weissman. 1987.Bone marrow stromal cell lines with lymphopoietic activity express high levelsof a pre-B neoplasia-associated molecule.Cell 48:1009.
31. Wineman, J., K. Moore, I. Lemischka, and C. Muller Sieburg. 1996. Functionalheterogeneity of the hematopoietic microenvironment: rare stromal elementsmaintain long-term repopulating stem cells.Blood 87:4082.
32. Marshall, A. J., H. E. Fleming, G. E. Wu, and C. J. Paige. 1998. Modulation ofthe IL-7 dose-response threshold during pro-B cell differentiation is dependent onpre-B cell receptor expression.J. Immunol. 161:6038.
33. Schlissel, M. S., L. M. Corcoran, and D. Baltimore. 1991. Virus-transformedpre-B cells show ordered activation but not inactivation of immunoglobulin generearrangement and transcription.J. Exp. Med. 173:711.
34. Carsetti, R. 2000. The development of B cells in the bone marrow is controlledby the balance between cell-autonomous mechanisms and signals from the mi-croenvironment.J. Exp. Med. 191:5.
35. Bain, G., E. C. Robanus Maandag, H. P. te Riele, A. J. Feeney, A. Sheehy,M. Schlissel, S. A. Shinton, R. R. Hardy, and C. Murre. 1997. Both E12 and E47allow commitment to the B cell lineage.Immunity 6:145.
36. Nutt, S. L., B. Heavey, A. G. Rolink, and M. Busslinger. 1999. Commitment tothe B-lymphoid lineage depends on the transcription factor Pax5.Nature 401:556.
37. Nutt, S. L., P. Urbanek, A. Rolink, and M. Busslinger. 1997. Essential functionsof Pax5 (BSAP) in pro-B cell development: difference between fetal and adult Blymphopoiesis and reduced V-to-DJ recombination at the IgH locus.Genes Dev.11:476.
38. Kozmik, Z., S. Wang, P. Dorfler, B. Adams, and M. Busslinger. 1992. The pro-moter of the CD19 gene is a target for the B-cell-specific transcription factorBSAP.Mol. Cell. Biol. 12:2662.
39. Kee, B. L., and C. Murre. 1998. Induction of early B cell factor (EBF) andmultiple B lineage genes by the basic helix-loop-helix transcription factor E12.J. Exp. Med. 188:699.
40. Nutt, S. L., A. M. Morrison, P. Dorfler, A. Rolink, and M. Busslinger. 1998.Identification of BSAP (Pax-5) target genes in early B-cell development by loss-and gain-of-function experiments.EMBO J. 17:2319.
41. Borge, O. J., J. Adolfsson, and S. Jacobsen. 1999. Lymphoid-restricted develop-ment from multipotent candidate murine stem cells: distinct and complimentaryfunctions of the c-kit and flt3 ligands.Blood 94:3781.
42. Fulop, G. M., and R. A. Phillips. 1989. Use of scid mice to identify and quantitatelymphoid-restricted stem cells in long-term bone marrow cultures.Blood 74:1537.
43. Rolink, A. G., S. L. Nutt, F. Melchers, and M. Busslinger. 1999. Long-term invivo reconstitution of T-cell development by Pax5-deficient B-cell progenitors.Nature 401:603.
44. Rebel, V. I., C. L. Miller, G. R. Thornbury, W. H. Dragowska, C. J. Eaves, andP. M. Lansdorp. 1996. A comparison of long-term repopulating hematopoieticstem cells in fetal liver and adult bone marrow from the mouse.Exp. Hematol.24:638.
45. Scollay, R., J. Smith, and V. Stauffer. 1986. Dynamics of early T cells: prothy-mocyte migration and proliferation in the adult mouse thymus.Immunol. Rev.91:129.
46. Rodewald, H. R., K. Kretzschmar, S. Takeda, C. Hohl, and M. Dessing. 1994.Identification of pro-thymocytes in murine fetal blood: T lineage commitment canprecede thymus colonization.EMBO J. 13:4229.
47. Aurrand-Lions, M., F. Galland, H. Bazin, V. M. Zakharyev, B. A. Imhof, andP. Naquet. 1996. Vanin-1, a novel GPI-linked perivascular molecule involved inthymus homing.Immunity 5:391.
48. Kawamoto, H., K. Ohmura, and Y. Katsura. 1998. Presence of progenitors re-stricted to T, B, or myeloid lineage, but absence of multipotent stem cells, in themurine fetal thymus.J. Immunol. 161:3799.
49. Hardy, R. R., C. E. Carmack, S. A. Shinton, J. D. Kemp, and K. Hayakawa. 1991.Resolution and characterization of pro-B and pre-pro-B cell stages in normalmouse bone marrow.J. Exp. Med. 173:1213.
50. Hunte, B. E., M. Capone, A. Zlotnik, D. Rennick, and T. A. Moore. 1998. Ac-quisition of CD24 expression by Lin2CD431B220lowckithi cells coincides withcommitment to the B cell lineage.Eur. J. Immunol. 28:3850.
51. Ray, R. J., A. Stoddart, J. L. Pennycook, H. O. Huner, C. Furlonger, G. E. Wu,and C. J. Paige. 1998. Stromal cell-independent maturation of IL-7-responsivepro-B cells.J. Immunol. 160:5886.
52. Georgopoulos, K., M. Bigby, J. H. Wang, A. Molnar, P. Wu, S. Winandy, andA. Sharpe. 1994. The Ikaros gene is required for the development of all lymphoidlineages.Cell 79:143.
53. Urbanek, P., Z. Q. Wang, I. Fetka, E. F. Wagner, and M. Busslinger. 1994.Complete block of early B cell differentiation and altered patterning of the pos-terior midbrain in mice lacking Pax5/BSAP.Cell 79:901.
54. Zhuang, Y., P. Soriano, and H. Weintraub. 1994. The helix-loop-helix gene E2Ais required for B cell formation.Cell 79:875.
55. Lin, H., and R. Grosschedl. 1995. Failure of B-cell differentiation in mice lackingthe transcription factor EBF.Nature 376:263.
56. Veiby, O. P., A. A. Mikhail, and H. R. Snodgrass. 1997. Growth factors andhematopoietic stem cells.Hematol. Oncol. Clin. North Am. 11:1173.
57. Metcalf, D. 1998. Lineage commitment and maturation in hematopoietic cells:the case for extrinsic regulation.Blood 92:345.
58. Enver, T., C. M. Heyworth, and T. M. Dexter. 1998. Do stem cells play dice?Blood 92:348.
59. Kee, B. L., and C. J. Paige. 1996. In vitro tracking of IL-7 responsiveness andgene expression during commitment of bipotent B-cell/macrophage progenitors.Curr. Biol. 6:1159.
60. Tarlinton, D. M., L. M. Corcoran, and A. Strasser. 1997. Continued differentia-tion during B lymphopoiesis requires signals in addition to cell survival.Int.Immunol. 9:1481.
61. Inaba, K., M. Inaba, M. Deguchi, K. Hagi, R. Yasumizu, S. Ikehara,S. Muramatsu, and R. M. Steinman. 1993. Granulocytes, macrophages, and den-dritic cells arise from a common major histocompatibility complex class II-neg-ative progenitor in mouse bone marrow.Proc. Natl. Acad. Sci. USA 90:3038.
62. McDonald, T. P., and P. S. Sullivan. 1993. Megakaryocytic and erythrocytic celllines share a common precursor cell.Exp. Hematol. 21:1316.
63. Choi, K., M. Kennedy, A. Kazarov, J. C. Papadimitriou, and G. Keller. 1998. Acommon precursor for hematopoietic and endothelial cells.Development125:725.
![PAX5-KIAA1549L: a novel fusion gene in a case of pediatric B-cell … · 2017. 8. 28. · [19, 20]. In this work we describe a novel PAX5 rearrange-ment in a case of pediatric BCP-ALL](https://static.documents.pub/doc/80x56/60ffd20923ebf366791e509b/pax5-kiaa1549l-a-novel-fusion-gene-in-a-case-of-pediatric-b-cell-2017-8-28.jpg)
