Characterization and Functional Analysis of Adult Human Bone Marrow Cell Subsets in Relation to B-Lymphoid Development

By S. Pontvert-Delucq, J. Breton-Gorius, C. Schmitt, C. Baillou, J. Guichard, A. Najman, and F.M. Lemoine

To study the frontiers between pluripotent stem cells and committed progenitors and to further define the B-cell pathway in adult bone marrow (BM), CD34+ subpopula- tions and CD34- B-lineage cells were analyzed by multi- parameter flow cytometry, studied by light and electron microscopy, and in short-term and long-term cultures (LTC). While the total CD34+ cells represent 4.9% f 0.8 of BM mononuclear cells within the lymphoid-blast win- dow.73.8 f 3.5%. 14.4 2 1.8%and8.8 f 2.9%ofthem were CD34+ CD10- CD19-, CD34+ CD10+ CD19+, and CD34+ CDI O+ CDI 9-, respectively. CD34+ CDI O+ CD19+ cells represent a small homogeneous TdT+ cfi - blast population. Although expressing CD38 and high level

EMATOPOIESIS is sustained in human bone H marrow by a small compartment of hematopoietic stem cells (HSC) with extensive proliferative capacities. This very rare population that expresses the CD34 antigen (Ag)' can be highly enriched by many techniques as flow cytometry using CD34 monoclonal antibody (MoAb) alone or in combination with other markers such as HLA-DR, CD38, rhodamine 123 (Rho), or T h ~ l . * - ~ Thus, it turns out that the hematopoietic progenitor cells (HPC) compart- ment appears in fact very heterogeneous and encompasses various subpopulations that can be distinguished by their differences both in the expression of various Ag and in their growth potential using in vitro and/or in vivo Thus, HPC encompass (1) a cellular fraction CD34+ CD38++ HLA-DRh@' and Rhobnght,39427 which gives rise in methylcellulose short-term assay to myeloid colonies and (2) a more primitive fraction of cells CD34++ CD38- HLA- DR'O" and Rhodu" 4,6 capable of initiating long-term bone marrow cultures (LTC). Such LTC initiating cells (LTC-IC) have been defined by their ability to generate myeloid clon- ogenic cells for 5 to 8 weeks and may represent more primi- tive cells capable of reconstituting hematopoiesis in However, CD34 Ag is also present on a subpopulation of cells coexpressing CD 19 and CD 10.' The CD 19 membrane protein, a member of the immunoglobulin (Ig) superfamily, is a B-cell specific marker expressed at each developmental stage except that of the terminally differentiated plasma cell,g2Lo whereas CD 10, a membrane-associated neutral en- dopeptidase, which might be involved in the regulation of the earliest stages of stromal cell-dependent B lymphopoie- sis,11,12 is expressed on a subset of immature B cells but also on neutrophils and on nonhematopoietic cell type^.'^,'^ Al- though cells coexpressing CD34, CDlO, and CD19 Ags might represent the earliest recognizable B-lineage cells, ie, B-lymphoid progenitor cells, very little is known about their physiologic functions. This might be due on the one hand to the fact that most of the studies on B-cell development have been performed either on leukemic B cells or on fetal tis- sues, I0,15-18 and on the other hand to the difficulties to obtain specific assays for pluripotent stem cells and B-lymphoid progenitors.

In this report, our goal was to establish the phenotypic patterns of the various CD34+ subpopulations in order to

of HLA-DR antigens, like myeloid committed progenitors, they did not generate LTC, myeloid, and T lymphoid colo- nies suggesting that the CD34+ CDI O + CD19+ popula- tion represents exclusively B lymphoid committed progeni- tors. By contrast, all myeloid progenitors and LTC-initiating cells were found in the CD34+ CD10- CD19- cell frac- tion. This fraction appeared more heterogeneous and con- tained CD38- HLA-DR'" small cells, larger blasts, and promonocyte-like cells exhibiting small peroxidase-posi- tive granules. Interestingly, CDlO was also present on CD34+ CD19- cells. This population mainly coexpressed CD33 and gave rise to macrophagic colonies. 0 1993 by The American Society of Hematology.

better define the frontiers between pluripotent and lineage committed progenitors and to particularly delineate the early steps of the B lymphoid pathway in adult normal bone marrow. The overall data show that in normal adult bone marrow (1 ) CD34' CDlO- CD19- cells contain myeloid progenitors and LTC-initiating cells; (2) CD34+ CDlO+ CD19- cells only give rise to macrophagic colonies and mainly express CD33 antigen; (3) CD34' CDIO' CD19+ cells, albeit sharing with other CD34+ progenitors many phe- notypic features besides B-specific markers, have lost their ability to generate myeloid cells and may exclusively repre- sent B-lymphoid committed progenitors.

MATERIALS AND METHODS

Cells Human heparinized bone marrow was obtained after informed

consent either from fragments of femur bones removed during or- thopedic surgery from patients free of hematologic diseases or by iliac crest aspiration from normal volunteer donors for allogenic transplantation. Bone marrow cells from orthopedic surgery were first centrifuged at 200g for 10 minutes to remove the fat. Bone marrow mononuclear cells (BMMNC) were then isolated by ficoll- hypaque gradient (density 1.077 g/mL). Cells were washed and then stored overnight at lo7 cells/mL in phosphate buffered saline (PBS)

From the Department of Hematology, Facultk de Mkdecine Saint Antoine, Paris, France; the H6pital Henri Mondor, Crkteil, France; and the H6pital de la PitibSalp6trit?re, Paris, France.

Submitted June 18, 1992; accepted March 2, 1993. F.M.L. is supported by a grant (no. 6560)from Association pour

la Recherche sur le Cancer (ARC), VillejuiJ France and by a grant from the Ligue Nationale Francaise contre le Cancer. S.P.-D. is a recipient of a C.I.F.R.E. fellowship from the French ministry of re- search and technology and from Biosys company (Compit?gne, France).

Address reprint requests to F.M. Lemoine, MD. PhD. Depart- ment of Hematology, Facultk de Mkdecine Saint-Antoine, 2 7, rue de Chalignv, 7501 2 Paris, France.

The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. section I734 solely to indicate this fact. 0 1993 by The American Society of Hematology. 0006-49 71/93/8202-0032$3.00/0

Blood, Vol 82, No 2 (July 15), 1993: pp 41 7-429 41 7

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

418 PONTVERT-DELUCQ ET AL

+ 20% fetal calf serum (FCS). Under these conditions, 87.6 k 7.3% of viable cells were recovered the following day.

Monoclonal Antibodies (MoAbs) and Fluorescent Reagents CDlO (J5), Phycoerythrin (PE) conjugated CDlO (J5-PE) and

CD33 (My 9-PE), biotinylated CD19 (B4-BIOT) MoAbs, and iso- matched mouse (Ms) Ig were purchased from Coulter (Coultronics, Margency, France). CD38-PE and HLA-DR-PE antibodies were purchased from Becton Dickinson (Sunnyvale, CA). Fluorescein conjugated (FITC) CD34 (8G-12 FITC)I9 was a kind gift from Dr P.M. Lansdorp (Terry Fox Laboratory, Vancouver, Canada). Anti- human FITC conjugated Ig M ( p chain specific) and anti-human FITC conjugated Ig A (a chain specific) were obtained from Tag0 (Burlingame, CA). Streptavidin allophycocyanin (APC) and goat anti-mouse APC were purchased from Molecular Probes (Eugene, OR). Normal Ms Ig were obtained from Biosys (Compi?gne, France).

Cell Surface Staining For triple staining with CD34, CDIO, and CD19 MoAbs,

BMMNC at IO’ cells/mL were first stained in PBS 2% FCS with CD19-BIOT MoAb ( 5 pL/106 cells) for 20 minutes at 4”C, washed, and then stained with CD34-FITC (0.5 pg/106 cells), CDIO-PE (5 pL/106 cells) MoAbs, and APC ( 1 pL/106 cells) for 20 minutes at 4 T . Labeled cells were kept on ice until analysis and sorting. In some cases, HLA-DR-PE, CD33-PE, or CD38-PE (20 pL/ IO6 cells) MoAbs were used instead of CDIO-PE. For triple staining with CD34, CD38, CDlO MoAbs, or with CD34, HLA-DR, CDlO MoAbs or with CD34, CD33, CDlO MoAbs, cells were first incu- bated with CDlO ( 5 pL/106 cells), then with GAM-APC (5 pL/106 cells). Before adding the other combinations of MoAbs, Ms Ig (17 pg/ IO6 cells) was used to block any free binding sites on GAM-APC. As negative control, cells were stained with conjugated isotype matched nonspecific Ms Ig and APC or GAM-APC.

Flow Cytometry Flow cytometry analysis and cell sorting were performed on a two

air cooled laser cytometer EPICS Elite (Coultronics, Margency, France). Fluorescence attributable to FITC and PE labeled MoAbs was determined using excitation by an argon laser operating at 488 nm and adjusted at 15 mW. APC was excited with an helium neon laser operating at 633 nm and adjusted at 10 mW. Forward and side scatter measurements were made using linear amplification and all fluorescence measurements were made with logarithmic amplifica- tion. Cells labeled with individual antibodies were used to set the voltage, and gain setting and compensation levels on the flow cy- tometer. For analysis and sort, a gate was drawn according to a low to moderate forward scatter (FS) and to a very low side scatter (SS) (see Results). Within this gate, the proportion of positive cells was estimated by determining first the channel number where the nega- tive control and the test sample curvescrossed each other. For analy- sis on triple stained cells, a minimum of 180,000 ungated events or 30,000 events gated on CD34+ cells were acquired in listmode using a software Elite, which permits the identification of multiple cell populations using multidimensional analysis. In some cases, simul- taneous expression of four antigens was monitored using sequential combination with three MoAbs and then deduced by comparing the results for each triple labeling procedure with another (see re- sults).

Cells were sorted at 2,000 cells per second and collected in PBS containing 50% FCS. The cell viability evaluated by trypan blue exclusion was always 95%. The purity, which attained 9670, was

assessed by analyzing immediately after sorting an aliquot of the sorted cells.

Intracellular Staining For intracellular staining, sorted cells were cytocentrifuged 12

minutes at 750 rpm. For intracytoplasmic p chain (CF) staining, slides were kept either at room temperature or at -20°C for several days before staining in order to decrease the background due to the remaining fluorescence present on the sorted cells. Both techniques gave similar results. Then, slides were fixed 10 minutes in cold methanol, washed three times in PBS, saturated 10 minutes in PBS 1% BSA, and incubated 40 minutes in a humid dark atmosphere with anti-human p chain-FITC or a chain-FITC (as negative con- trol) antibodies at a final dilution of 1/100. After three more washes, slides were mounted and 200 cells per slide were counted with a fluorescence microscope. The percent surface p chain posi- tive cells was determined by subtracting the percentage of a chain positive (nonspecific staining <3%) from the percentage of p chain positive. As positive control, an EBV p chain positive cell line gen- erated in our laboratory was used.

Terminal deoxynucleotidyl transferase (TdT) was detected using an immunoperoxidase kit obtained from Supertechs Inc (Bethesda, MD).” Briefly, unfixed slides were stored in a dessicator at room temperature before use, then slides were fixed 30 minutes in abso- lute methanol at 4”C, air dried, and stained according to the Super- techs’ procedure. Slides were mounted and counted by bright field microscopy. As positive controls, Supertechs slides and a B ALL TdT positive from our laboratory were used. As negative control, the rabbit anti-TdT antibody was replaced with PBS (0% ofthe cells were positive).

Preparation for Light and Electron Microscopy For light microscopy, sorted cells (punty 296%) were cytocentri-

fuged as described above, fixed, and stained with a May Griinwald Giemsa coloration. At least 200 cells per slide were examined by two independent cytologists. For electron microscopy, a modified micromethod from Anderson” was used. First, 5,000 to 10,000 cells per fraction were sorted into 3% ofglutaraldehyde in 0.1 mol/L phosphate buffer, incubated 1 hour, centrifuged, washed, fixed in 1% osmic acid solution, washed again in PBS, and placed in Beem capsules. Capsules were centrifuged, and 5 to 10 mL of human plasma and 3% of glutaraldehyde in 0.1 mol/L phosphate buffer were gently added to the pelleted cells. After 1 hour incubation, cells were dehydrated and embedded in epon. This method allowed to manipulate a solid block in which an invisible pellet of cells was lodged. For the detection of peroxidase activities, cells fixed by glu- taraldehyde were incubated for 1 hour in Graham Kamovsky me- dium’’ and treated as above. The thin stained sections were exam- ined using a Philips CM 10 electron microscope.

Methylcellulose Colony Assay .for Myeloid Progenitors Cells were plated in 35 mm dishes (Greiner, FRG), in 1. I mL of

IMDM containing 0.8% methylcellulose (Fluka 4000, Buchs, Swit- zerland), 30% FCS, 1% BSA, 100 pmol/L 2-mercaptoethanol (Sigma), 2 mmol/L glutamine, 100 U/mL penicillin, 100 pg/mL streptomycin, 3 U/mL recombinant human erythropoietin (rh Epo) (Boehringer, Mannhein, FRG), 100 U/mL rh Interleukin 3 (IL3) (Genzyme, Boston, MA), 200 U/mL rh granulomacrophagic colony stimulating factor (GM-CSF) (Genetics Institute, Cam- bridge, MA).

Duplicate cultures were incubated in a humidified atmosphere containing 5% COz in air at 37°C. BMMNC were plated at 5 X IO4 per dish, and sorted fractions were plated at various cell concentra-

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

STUDY OF CD34' SUBSETS AND B-CELL PRECURSORS 41 9

tions (IO' to IO5 cells per dish) depending of their phenotype. Burst forming unit erythroid (BFU-E) and colony forming unit granulo- macrophagic (CFU-GM) were counted between 16 and 18 days culture. BFU-E were subdivided into primitive and mature sub- classes according to their number of cluster~.~' CFU-GM included CFU-GM, CFU-G, and CFU-M, but in some experiments, CFU- GM colonies were subdivided into CFU-G/GM and CFU-M (see Results).

Long- Term Marrow Cultures LTC were established and maintained as described elsewhere.2426

Briefly, cells were suspended in 2.5 mL of long-term culture me- dium (Terry Fox Laboratory media, Vancouver, Canada), supple- mented with mol/L hydrocortisone hemisuccinate (Sigma, St Louis, MO) in 35 mm tissue culture dishes (Falcon, Becton Dickin- son, Plymouth, UK) on a preestablished irradiated ( 1 5 Gy) normal feeder layer obtained from allogeneic marrow as follows: LTCs were initiated 3 to 6 weeks before the experiment, adherent layers were trypsinized, irradiated with 15 Gys, and replated in 35 mm tissue culture dishes at 3 X IO4 cells/cm2. On these feeder layers, BMMNC were plated at lo6 cells/LTC and sorted fractions at dif- ferent concentrations depending of their phenotypes. Cultures were first incubated 3 days at 37°C and thereafter at 33°C in 5% COz. Weekly, half of the nonadherent cells were removed and half of the medium was replaced. Five weeks later, nonadherent and adherent cells obtained after trypsinization of the adherent layer were pooled and tested for clonogenic activity.24 This provides an indirect mea- surement of the number of LTC-IC present at day 0 as desc~ibed.~.'

Statistical Analysis

(SEM). Results were presented as the mean i standard error ofthe mean

RESULTS

Immunophenotyping Patterns of CD34+ Cells and B-Cell Precursors

Light scatter properties. The study of the light scatter properties of CD34' cells and B-cell precursors indicate that they are contained in a window, so called lymphoid blast gate representing 50.9 k 2.55% of the total BMMNC (Fig 1A) and already known to contain all the myeloid progeni- tors, ie, clonogenic and LTC-IC (unpublished data).7 Within the CD34' cells, CD34' CDlO- CD19- cells and CD34+ CD 1 O+ CD 19- cells have a low to intermediate FS and a very low SS (Fig 1B and C), while CD34' CDlO' CD 19+ cells (Fig 1 D) are somewhat smaller (lower FS) and more homogeneous. Interestingly, CD34- cells expressing B cell antigens, ie, CD 1 O+ CD 19' or CD 19+ alone have, like CD34+ CD 1 O+ CD 19+ cells, very low FS and SS (Fig 1 E and F). Data from 2 1 separate experiments indicate that within the lymphoid blast gate, CD34+ CDlO- CD19-, CD34' CD10' CD19+, CD34+ CDIO' CD19-, CD34- CDlO+ CD19+, and CD34-CD10- CD19+ fractionsrepresent 3.7% +0.7%,0.7%+0.2%,0.3%+0.04%,5.1%+0.8%,and8.5% + 0.9% of the BMMNC, respectively.

Distribution of CD34, CDIO, and CD19 antigens in the lymphoid blast gate. The fluorescence profile of BMMNC stained with CD34, CDlO, and CD19 conjugated MoAbs is presented in Fig 2A through C. The mean percentage (n = 2 1 ) of CD34' cells, CD 1 O+ cells, and CD 19' cells within the

lymphoid-blast gate is 4.9% t 0.8%, 7.5% -+ 1.1%, and 14.4% -t 1.2%, respectively.

The analyses of the distribution of CD I O and CD 19 on CD34+ gated cells (Fig 2D and Fig 3A) indicate that CD34+ CD 10+ CD 19' cells and CD34' CD 1 O+ CD 19- cells repre- sent 14.4% and 8.8% of the total CD34' cells, respectively, whereas no CD34+ CDlO- CD19' cells are detected. This strongly suggests that, in adult bone marrow, all CD34+ CD 19+ cells coexpress CD 10.

Analyses performed on CD10+ gated cells (Fig 3B) or on CD19' gated cells (Fig 3C) indicate that CD34+ CDIO' CD19' cells represent 10.7% and 4.4% of the total CDlO and CD19 cells, respectively, whereas CD34- CD 10+ CD19+ cells represent 67.2% and 35.3% of the total CDlO and CD19 cells, respectively. Interestingly, one can see (Fig 3B) that 22.1% of the CDlO+ cells do not coexpress the CD 19 B-cell-specific antigen.

Therefore, we studied whether CD34+ CDIO' CD19- cells might express CD33 usually present on myeloid cells.27 For this purpose we used sequential triple combinations with CD34, CDlO, CD19, and CD33 MoAbs. Analysis per- formed on 30,000 gated CD34' cells shows that CD34' CD19' cells are CD33- (Fig 2E), while 8.2% ofCD34' cells are CDlO' and CD33+ (Fig 2F). Thus, the comparison of the different triple stainings clearly indicates that CD34' CD10' CD19- cells (8.8% of the CD34+ cells) mainly ex- press the CD33 Ag. This suggests that CDlO might be pres- ent on myeloid progenitor cells.

Because CD38 and HLA-DR antigens are useful to distinguish very primitive CD34+ cells from myeloid clonogenic progeni- tors, we studied the expression of these two antigens on CD34+ and CD34- cells expressing B-lymphoid markers using triple staining procedure with combination of CD34, CD10, CD19, CD38, and HLA-DR MoAbs. Analyses within CD34+ cells indicate that CD34+ CD19+ are HLA- DR' (Fig 2G) and CD34+ CD19' are CD38+ (Fig 2H). As- suming from Fig 2D that CD34' CD19+ cells are always CD lo+, these data indicate that like clonogenic myeloid CD34+ progenitors, CD34' CDlO+ CD19+ cells are also CD38 and HLA-DR positive. Additional experiments show that, in the CD34+ CDlO- (also CD19-) cell fraction, 15% and 7.5% are HLA-DR'O" and CD38-, respectively (Table I). These cells, ie, CD34' HLA-DR'OW or CD34+ CD38- express high levels of CD34' Ag and represent a small-sized and homogeneous population (not shown).

Similar cytometric analysis were also performed on CDlO' gated cells or on CD19+ gated cells. The data are summarized in Table 1 and show that CD34- cells coex- pressing CD 10 and CD I9 antigens are CD38+ and HLA- DR+. Interestingly, CD38 antigen is only expressed in 20% of the CD34- CDlO- CD19+ cells, which represent newly formed and mature B cells.

In order to determine the expression of p chain and TdT on the different cell frac- tions, sorted cells were cytocentrifuged and stained as de- scribed in the Materials and Methods section. As shown in Fig 4, TdT is strongly expressed on CD34+ CDlO+ CD19+ cells and progressively decreases along the B lymphoid devel-

Expression of CD38 and HLA-DR antigens.

p Chain and TdT expression.

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

420

3-

1-

m.

v)

%-

A-

PONTVERT-DELUCQ ET AL

%-

1-

1-

LL 3- v)

m.

A TOTAL BMMNC

3

I

1

9

A

s

v)

I 9 Ib 2b XE 4b S b bb

ss

D CD34+ CD10+ CD19+

B CD34+ CD10- CD19-

'1

E CD34- CD10+ CD19+

C CD34+ CD10+ CD19-

F CD34- CD10- CD19+

Fig 1. Forward scatter (FS) and side scatter (SS) properties of CD34+ cells and B-cell precursors. A representative of 21 separate experiments is shown. (A) Total BMMNC and definition of lymphoid-blast window; (B) CD34+ CD10- CD19- cells; (C) CD34' CD10+ CD19- cells; (D) CD34+ CDlO+ CD19+ cells; (E) CD34- CD10+ CD19+ cells; (F) CD34- CD10- CD19+ cells.

opment, while intracytoplasmic p chain expression appears on cells that no longer display CD34. Interestingly, very few CD34' CDIO- CD19- cells are TdT positive. These data indicate that the majority of CD34' CD 10' CD 19' cells are TdT' and cp-.

Morphologic Aspect qf CD34' Sirbpopulations and CD34- B-Lineage Cells

Light microscopy. As described in the Materials and Methods section, at least 200 cells per slide were studied. The results show that CD34+ CD IO- CD 19- cells are com- posed of blast cells but are not homogeneous by their size (Fig 5A). Most ofthem exhibit one or two large nucleoli and in some cases the nonbasophilic archoplasm may contain some azureophilic granules (Fig 5B). CD34' CD I 0' CD 19' sorted cells also represent a heterogeneous population with some cells identical to CD34' CDIO- CDIY cells, while others have a very compact chromatin. Interestingly, as in the CD34' CDIO- CD19- subpopulation, mitotic cells are also observed (Fig 5C). The morphologic aspect of CD34- CDIO' CD19+ cells is illustrated in Fig 5D: some of them are of smaller size with a high nuclear cytoplasmic ratio. Surprisingly. in the CD34- CDIO- CD19' cell fraction, nu- cleoli are rather frequent in the nucleus (Fig 5E). Because of

the low number of CD34+ CDIO' CD19- cells, this cell fraction has not been examined.

Electron microscop?,. The morphologic aspect con- firmed the observations made by light microscopy. The dif- ferent cells presented in the Figs 6, 7, and 8 are representa- tive of seven separate cell sorts that gave identical results. Blasts from CD34' CD IO- CD 19- cell fraction show a nu- cleus with predominant euchromatin. The structure of the nucleoli is either loose (Fig 6A) or compact (Fig 6B). Mono- ribosomes are abundant but the endoplasmic reticulum cis- ternae are rare or absent (Fig 6A, B). Some of the CD34' CDIO- CD19- blasts exhibit few clustered granules and contain long cisternae ofendoplasmic reticulum and numer- ous mitochondria (Fig 6C). After incubation with diamino- benzidine medium, all the small granules are peroxidase- positive (inset Fig 6C). In contrast to the usual bone marrow promonocytes, the peroxidase activity is never detected in the nuclear envelope, endoplasmic reticulum, and Golgi ap- paratus. These results suggest that these blasts stop very early the production ofperoxidase-positive granules and rep- resent a subpopulation of CD34' promonocytes. Interest- ingly these cells are never found in the CD34' CDIO' CD19' cell fraction. Although CD34' CDIO' CD19' cells contain several blasts exhibiting a morphology similar to that shown in Fig 6A and B, the majority of them have a

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

STUDY OF CD34' SUBSETS AND B-CELL PRECURSORS 42 1

A

I-=

CD34-FITC

2 -3 I W

O m

0 0

4 F-

. I 1 le 108 ieei CD1 9-APC

B C

CD10-PE C D19-A PC

4 c)Q) m- 0 U

- . 1 1 ie

CD10-APC

Fig 2. Expression of CD10, CD19, CD33, CD38, and HLA-DR antigens within the lymphoid-blast window. Single fluorescence profile for CD34-FITC (A), CD1 0-PE (6). and CD1 9-APC (C). Analysis was performed on 180,000 gated events in the lymphoid-blast window. Three color analysis showing the expression of various antigens was evaluated by gating on 30,000 CD34+ cells. (D) Coexpression of CD1 0-PE and CD1 9-APC; (E) expression of CD33-PE and CD19-APC; (F) coexpression of CD33-PE and CDl O-APC; (G) coexpression of HLA-DR-PE and CD19-APC; (H) coexpression of CD38-PE and CD1 9-APC. All fluorescence profiles are presented in a logarithmic scale. A representa- tive of 21 and 5 separate experiments for (A) through (D) and for (E) through (H) is shown, respectively.

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

422

A

CD34+ 10- 19-

CD34+ 10+ 19+

CD34+ 10+ 19-

C D M 10+ 19+

CD34- 10+ 19-

CD34- 10- 19+

B

CD34+ 1Q 19-

CD34+ 10+ 19+

CD34+ 10+ 19-

CD34- 10+ 19+

CD34- 10+ 19-

CD34- 10- 19+

C

CD34+ 10- 19-

CD34+ 10+ 19+

CD34+ 10+ 1%

CD34- 10+ 19+

CD34- 10+ 19-

CD34- 10- 19+

PERCENT OF TOTAL CLW+ CELLS

8 8 8

14.4f1.8%

PERCENT OF TOTAL CD10+ CELLS

m 4.6i lK B

PERCENT OF TOTAL CD19+ CELLS

I 4.4M.Ph I

PONTVERT-DELUCQ ET AL

nucleus with more heterochromatin and nucleoli (Fig 7A, B). Other cells have a very irregular and indented nucleus (Fig 7C). Cells in mitosis (Fig 7D) are also observed. In the CD34- CD 1 0' CD 1 9' or CD34- CD IO- CD 19' fractions, the development of endoplasmic reticulum is more evident (Fig 8) and the loose or compact nucleoli persist (Fig 8A and B). As seen by light microscopy, a nucleolus also remains in the nucleus of many CD19' cells (Fig 8C). Interestingly by comparing Figure 8A and D taken at the same magnifica- tion, it turns out that the size of mitochondria clearly in- creases in mature B cells lacking CD 10 (Fig 8D). These mor- phologic data indicate that CD34+ CD 1 O+ CDl9' cells strongly resemble the other CD34+ subpopulations, but are a little bit more homogeneous.

Culture Assays To compare the growth abilities of the different CD34'

subpopulations and CD34- B-lineage cells, we used various culture assays capable of detecting myeloid progenitors. When sorted cell fractions are plated in myeloid short-term assay, we observe that myeloid and erythroid colonies (Ta- ble 2) are only present in the CD34' CD 10- CD 19- subset, while no colonies are detectable under our culture condi- tions in the CD34' CDlO+ CD19+ cell fraction. Interest- ingly, CD34' CD 10+ CD 19- cells are capable of giving rise to - 30% (40 of 130) of the macrophagic colonies within the total CD34' population, indicating that CDlO is expressed by progenitors that still retain at least some myeloid poten- tial.

To measure the number of LTC-IC contained in the dif- ferent sorted cell fractions, cells were plated in LTC as de- scribed in the Materials and Methods section. The results show in Table 3 that myeloid clonogenic cells detectable after 5 weeks culture, are only present in the CD34' CD 10- CD 19- cell fraction.

These data indicate that CD34+ CDIO- CD19- cells con- tain both clonogenic myeloid progenitors and LTC-IC, whereas CD34' CD 10' CD 19' cells have lost their myeloid potential and may consequently represent B-lymphoid committed progenitors.

DISCUSSION

The hematopoietic system is composed in the bone marrow of a cellular hierarchy where the most primitive cells, which possess a pluripotent potential, proliferate and differentiate into committed progenitors giving rise through multiple steps of differentiation to circulating blood cells. This continuous orderly process is characterized by coordi- nated changes both at the gene level and in the expression of intracellular and cell surface antigens. In this report we fo- cused our attention on early B-lymphoid progenitors in adult bone marrow. For this purpose, we sorted by multi- parameter flow cytometry the various CD34+ subpopula-

Fig 3. Distribution within the lymphoid blast window of the CD34+ subpopulations and CD34- B-lineage cells on CD34+ (A), CD1 O+ (B). and CD19' (C) gated cells. Results represent the mean f SEM of 21 separate experiments.

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

STUDY OF CD34' SUBSETS AND B-CELL PRECURSORS 423

Table 1. Percentage CD38 and HIA-DR Positive Cells Within CD34' Subpopulations and CD34- B-Lineage Cells

CD38' Cells (%) HLA-DR' Cells (%)

CD34'CDlO- CD19- 92.5 100' CD34+ CDlO+ CD19' 100 100 CD34- CD1 O+ CDl9+ 100 100 CD34- CD10- CD19' 20 100

The number of single, double, and triple positive cells using various combinations of three MoAbs were calculated using a software Elite. Then, the percentage of CD38 or HLA-DR positive cells were deduced by comparing the combinations of three MoAbs to one another. Results for CD38 and HLA-DR were obtained from three and five separate ex- periments, respectively.

In the CD34' CD10- CD19- cell fraction, 15% of the cells were HLA- DR"", whereas 85% were HLA-DRmh.

tions and CD34- B-lineage cells, and we compared their immunophenotypic, morphologic, and functional patterns. First, we analyzed the distribution of CD I O and CD 19 anti- gens on CD34' cells. While CD34+ CD 10' CD 19' cells and CD34+ CDIO' CD19- cells represented 14.4% and 8.8% of the total CD34+ cells, respectively, no CD34' CDIO- CD19' cells were detectable. These findings indicate that CD34' CD19' cells are always CDIO+ and suggest that CD 19 is not expressed before CD I O antigen in adult bone marrow. This is of particular interest because of the contro- versy concerning the sequential expression of CDIO and CD19 antigens. Indeed, based on the identification of CDIO- CD19' leukemic cells, Nadler et al have proposed that CD 19 expression precedes CD I O expression," whereas other studies camed out in fetal liver, where CD IO' CD 19- cells are in larger number than CD IO' CD 19' cells, indicate that CDIO antigen expression precedes the acquisition of

PERCENT OF msinvE CELLS A

0 0 CO 0 al 0 P 0 N 0 0

Fig 4. Terminal deoxynucleotidyl transferase (TdT) and intracy- toplasmic p chain (cp) expression in the CD34+ subpopulations and CD34- B-lineage cells. Results represent the mean 2 SEM of three separate experiments. (m) Percentage of TdT' cells; (0) per- centage of cp' cells.

B

Fig 5. Light microscopy of CD34+ subpopulations and CD34- E-lineage cells. Each sorted cell fraction was cytocentrifuged and stained by May GrUnwald Giemsa. (A and E) CD34' CD10- CD l9 - cells. All cells exhibit one or two large nucleoli. The nucleus is indented by the nonbasophilic archoplasm. (B) Some azureophilic granules in this zone are shown. (C) CD34' CD10' CD19+ cells. A large cell in mitosis is surrounded by cells with very dense nu- cleus. (D) CD34- CD10+ CD19+ cells. As in other fractions, the cell size is variable as well as the nuclear cytoplasmic ratio. (E) CD34- CD10- CD19+ cells. These cells have the morphology of a lymphocyte. However, note the presence of a nucleolus in the lower cell.

CD19 antigen." These apparent discrepancies might be due to the fact that most ofthe studies on human B lymphopoie- sis have been carried out either on fetal tissues (ie, fetal liver or fetal bone marrow) or on leukemic B cells.*' Therefore, we suggest that the antigenic distribution might be different in the fetal and adult B lymphoid pathway. This hypothesis

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

424 PONTVERT-DELUCQ ET AL

Fig 6. Ultrastructural aspect of CD34+ CD10- CD19- blast cells. (A) Aspect of a blast at low magnification ( X 5,000). The enlarge- ment shows a large loose nucleolus (Nu) surrounded by heterochromatin. In the cytoplasm, there are numerous monoribosomes ( X 25,000). (B) Low magnification [ X 5,000) of another smaller blast with an indented nucleus. At higher magnification ( X 25.000). the round nucleolus appears compact. (C) Blast with different characteristics. A large nucleolus is also present in the nucleus with an irregular contour. In contrast with the two other blastsfrom A and B, this blast exhibits in the cytoplasm long cistemae of endoplasmic reticulum (ER) and clustered small granules (Gr) ( X 12,200). (C, inset) Identical blast that has been incubated in the diaminobenzidine medium. The very small granules are peroxidase positive. Compare their size with that of mitochondria ( X 35,000).

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

STUDY OF CD34' SUBSETS AND B-CELL PRECURSORS 425

A - B

1 . e..

Fig 7. Ultrastructural aspect of CD34' CD1 O + CD19+ blast cells. (A) This blast with a loose nucleolus has a nudeus with het- erochromatin distributed mainly around the nuclear envelope and surrounding the nucleolus ( X 8.900). (B) Another blast with two nucleoli. A centriole (Ce) is in the middle of a Golgi zone, which deforms the nu- cleus ( X 8,900). (C) Larger blast with highly indented nu- cleus and small multiple nu- cleoli (arrows) ( X 8,900). (0) Cell in mitosis ( X 8,900).

&$&? ++#

. . . .

is reinforced by the fact that we did not find any CD34' CD 1 0' CD 19' cells in cord blood (our unpublished data) in line with others.29 Although our data did not determine whether some CDIO' CD19' cells may be committed to the B lineage, the isolation of various CD34' subpopulations should enable us to answer this question either by using the B-cell colony assay recently described by McGinnes et aI3O or by developing a human counterpart of the murine lym- phoid long-term ~ulture.~' . '~ Recently, such a culture assay in the presence of stromal cells and IL7 has been described for fetal human B-cell precursors."

Further characterization of CD34+ CD I 0' CDI 9+ cells indicate they were TdT' and cp-. Interestingly, we also found that 2% of CD34' CDIO- CD19- cells were TdT' (Fig 4). These results are in agreement with other data sug- gesting that TdT is either expressed on B cell progenitors before the acquisition of CD 19 and CD IO Ags or on CD34' cells that likely include progenitors cells for both B and T lineage." The concept of a common lymphoid progenitor is still unclear albeit a subpopulation of lymphoid progenitors

3

*.'

I I .

coexpressing CD 19, CD7, and CD34 has been identified in human fetal BM." Interestingly, we also found CD34' CD7' CD 19' cells and CD34' CD7' CD 19- cells in adult bone marrow (not shown). The demonstration that such a putative common lymphoid progenitor can differentiate both into phenotypically mature T and B cells would be proven by using B lymphoid3' and prothymocyte However, in preliminary experiments performed with CD34' CD IO- CD 19- and CD34+ CD 10' CD 19' subpopu- lations we did not get any T cell clones (data not shown). This negative result might be due to the extremely low fre- quency of CD34' CD7' cells, the most immature prethy- mic T lineage pr~geni tor ,~~ which only represent 5% of the total CD34+ cells and give rise in limiting dilution assay to T cell clones at a frequency of 1 in 35 to 53.40 In order to demonstrate whether CD34' CD 10' CD 19' cells still retain some T lymphoid potential, further experiments on purified CD34' CD7' CD I9+ cells that are also CD 1 0' are now in progress.

Morphologic and functional studies indicated that

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

426 PONTVERT-DELUCQ ET AL

A B *: ...

D

; ER

ER'

Fig 8. Uttrast~ctural aspect of CD34- CD1 O+ CD19+ blast cells (A and 6) and CD34- CD10- CD19+ (C and 0) cells. (A) A compact nucleolus is pres- ent in the nucleus. Note the high nuclear cytoplasmic ratio and the small size of mitochon- dria ( X 8,900). (6) large blast with irregular nucleus with a large nucleolus ( X 8,900). (C) Cell exhibiting a nucleolus in a nucleus with abundant hetero- chromatin ( X 8,900). (D) A small lymphocyte. Note the large size of mitochondria com- pared with that in (A). Several cistemae of endoplasmic reticu- lum are seen.

CD34' CDlO' CD19' cells had a blast morphology like other CD34+ cells but were more homogeneous and of a smaller size, and were not capable of growing in short-term clonogenic myeloid culture nor in LTC. This loss of my-

eloid potential is in agreement with previous studies on hu- man fetal hematopoietic tissues showing that purified CD 10' lymphoid progenitors were not capable ofgiving rise to CFU-GM colonies.4' However, the presence of mitosis

Table 2. Detection of the Number of Myeloid Clonogenic Progenitors in the Different Subpopulations

8FU-E Total CFU-G/M No. of

CFU-M Experiments CFU-G/M* CFU-G/GM 3-8 Clusters 8-1 6 Clusters > 16 Clusters

Mononuclear cells

Total CD34+ Total CD34-

CD34+ CD10- CD19-

CD34+ CD10* CD19+ CD34+ CD10+ CD19- CD34- CD10+ CD19+ CD34- CD10- CD19+

7 2 1 5.8 f 1.5 1 3 2 4

63 f 48 63 k 58 56.5 f 41.5 1.52 1 0.3 f 0.3 0.6 k 0.1

114 f 20.6 95 f 25 216 f 69

0 I f 1 1 f 1 0 0 0 0 0 0 0 0 0

33.5 2 4

2 7 0 f 123 7.5 2 2.3

272 f 40

3.5 f 2.3 4 0 f 18

0 0

8 22.5 f 1.5 3.4 2 1.9 2 140 k 91 130 2 108 3 3.3 2 1 4 f 3 3

9 155 f 98 76.5 f 53 3

0 3.5 f 2.3 3 0 4 0 2 18 3

8 8

Absolute number of progenitors are given for 10,OOO cells. Results represent the mean f SEM of (n) separate experiments performed in duplicate. CFU-G/M included CFU-GM, CFU-G, and CFU-M.

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

STUDY OF CD34' SUBSETS AND B-CELL PRECURSORS 427

Table 3. Recovery of Committed Progenitors From Sorted Cell Fractions After 5 Weeks in Long-Term Studies

Total BFU-E Total CFU-GM

Total mononuclear cells 131 ? 17 279.5 ? 28.5 CD34+ CD10- CD19- 4,640 ? 740 9,300 t 0 CD34+ CD10+ CD19+ 0 0 CD34- CD1 Of CD 19+ 0 0 CD34- CD10- CDI 9' 0 0

One representative experiment of four separate experiments is shown. Sorted cell fractions were seeded in LTC. After 5 weeks, nonad- herent and adherent cells were recovered and plated in duplicate in methylcellulose. Absolute number of progenitors are given for 10' plated cells in LTC.

(Figs 5C and 7D), indicates that this population is not quies- cent and might grow under appropriate culture conditions.

Second, we studied CD34' CDlO- CD19- cells (ie, 73%- 74% of total CD34' cells), which appeared phenotypically more heterogeneous. Indeed, some of them contained clus- tered granules and were peroxidase-positive (Fig 6C). Sev- eral arguments indicate that these cells are indeed CD34' promonocytes and do not represent a contaminant: they are numerous in all CD34' CD10- CD19- fractions and are absent from all CD34' CD 1 Of CD 19' fractions studied; these cells have also been described in two other studies on purified CD34' and express early (CD13, CD33) and late (CD 14, CD 15) myeloid markers?3 In addition, us- ing immunogold labeling on a suspension of unseparated bone marrow cells, such promonocytes have also been iden- tified with anti-CD34 antibody (JBG, personal data). These cells probably are maturing into macrophages in vitro. Other CD34' CDlO- CD19- cells were smaller with com- pact nucleoli (Fig 6B). The latter cells, which present a minor component of CD34+ cells are CD38- and HLA-DR- low and express higher levels of CD34 antigen. Such cells, capable of initiating long-term bone marrow culture in vi- t1-0,~~' have an aspect similar to the murine WGA+ Rh123dU'1 cells44 and might be identical to the CD34' Thyl" cells re- cently described as a human hematopoietic pluripotent stem cell population.' Interestingly, when we examined the expression of the CD38 and HLA-DR Ags on CD34' CD 10' CD 19+ cells, we found that these cells were positive for both Ags like myeloid committed progenitors within the CD34+ CD10- CD19- subset. These findings indicate that CD38 and HLA-DR antigens are expressed coordinately with lineage specific antigens on CD34' cells and represent general markers of commitment in hematopoiesis.

Third, we observed that 8.8% of CD34+ cells were CD34' CD10+ CD19-. Further studies indicated that these cells were mainly CD33' (see Fig 2F), but such a cell population has not been found by others.29 Interestingly, CD34' CDlO' CD19- cells, when plated in culture, gave rise to macrophagic colonies. From these findings, one can specu- late about the existence of a common progenitor to the mac- rophagic and lymphoid lineages in human bone marrow inasmuch as recent published data have shown the presence of bipotential precursors of B cells and macrophages in mu- rine fetal l i ~ e r . ~ ' In view of our results, we propose that the

hierarchical organization of the CD34' subpopulations in adult bone marrow would be as follows: 1) CD34' CD38- HLA-DRIoW cells would represent the most primitive pluri- potent stem cell; 2 ) then this small sized population (with a morphology similar to Fig 6B) begins to be committed and to express CD38 and a high level of HLA-DR Ags; 3) within this population of CD34' CD38' HLA-DRhigh cells, one can suppose that some of them commit to the myeloid lineage and others represent a common progenitor to the macro- phagic and lymphoid lineages. The latter cells, which still retain some ability to generate macrophagic colonies, are CD34+ CDlO+ and might coexpress either CD33 and/or T and B associated markers including CD7, CD19, and TdT; 4) finally, CD34' CD 1 O+ CD 19' cells, which have lost their myeloid potential, may represent the earliest B-lymphoid committed progenitor capable of differentiating into CD34- CD 10' CD 19' cpc pre-B cells and newly formed CD34- CD 10- CD 19+ B cells.

Therefore, the characterization of these different stages of the B lymphoid pathway from the most immature CD34+ cells to the mature B lymphocytes will allow us to study their response to growth factors such as interleukin-7 and kit-ligand, which have been shown to stimulate, in combina- tion with stromal cells, the proliferation of pre-B cells in mice.46 Such a strategy, might provide the appropriate cul- ture conditions to develop B-lymphoid specific assays in human adult bone marrow.

ACKNOWLEDGMENT

We thank ARC, Beecham Institute, the Groupe des Assurances Nationales (CAN), and Musica Ltd for their financial contribution to the cell sorter EPICS Elite (Coultronics, France). We are also grateful to Prof A. Apoil, Prof T. Judet, Prof E. Gluckman, and Dr R. Traineau for providing bone marrow samples and to Dr P.M. Lansdorp (Terry Fox Laboratory, Vancouver, Canada) for the kind gift of 8G12-FITC monoclonal antibody.

REFERENCES 1. Civin CI, Strauss LC, Brovall C, Fackler MJ, Schwartz JF,

Sharper IH: Antigenic analysis of hematopoiesis 111: A hematopoi- etic progenitor cell surface antigen defined by a monoclonal anti- body raised against KG-la cells. J Immunol 133:157, 1984

2. Lu L, Walker D, Broxmeyer HE, Hoffman R, Hu W, Walker E Characterization of adult human marrow hematopoietic progen- itors highly enriched by two-color cell sorting with My-IO and ma- jor histocompatibility class I1 monoclonal antibodies. J Immunol 139:1823, 1987

3. Terstappen LWM, Huang S, Safford M, Lansdorp PM, Loken MR Sequential generations of hematopoietic colonies derived from single non lineage committed CD34+ CD38- progenitor cells. Blood 77:1218, 1991

4. Udomsakdi C, Eaves CJ, Sutherland HJ, Lansdorp PM. Sepa- ration of functionally distinct subpopulations of primitive human hematopoietic cells using Rhodamine- 123. Exp Hematol 19:338, 1991

5. Baum CM, Weissman IL, Tsukamoto AS, Buckle AM, Peault B: Isolation of a candidate human hematopoietic stem-cell popula- tion. Proc Natl Acad Sci USA 89:2804, 1992

6. Sutherland HJ, Lansdorp PM, Henkelman DH, Eaves AC: Functional characterization of individual human hematopoietic

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

428 PONTVERT-DELUCQ ET AL

stem cells cultured at limiting dilution on supportive marrow stro- mal layers. Proc Natl Acad Sci USA 87:3584, 1990

7. Sutherland HJ, Eaves CJ, Eaves AC, Dragowska W, Lansdorp PM: Characterization and partial purification of human marrow cells capable of initiating long term hematopoiesis in vitro. Blood 74:1563, 1989

8. Loken MR, Shah VO, Dattilio KL, Civin CI: Flow cytometric analysis of human bone marrow. 11. Normal B lymphocyte develop- ment. Blood 70: 13 16, 1987

9. Nadler LM, Anderson KC, Marti G, Bates M, Park E, Daley JF, Schlossman SF: B4, a human B lymphoid-associated antigen expressed on normal, mitogen-activated and malignant B lympho- cytes. J Immunol 131:244, 1983

IO. Nadler LM: B cell/leukemia panel workshop: Summary and comments, in Reinherz EL, Haynes BF, Nadler BF, Bernstein ID (eds): Leucocyte Typing 11, vol 2: Human B Lymphocytes. New York, NY, Springer-Verlag, 1986, p 3

1 1. Letarte M, Vera S, Trans S, Addis J, Onizuka RJ, Quacken- bush EJ, Jongeneel CV, McInnes RR: Common acute lymphocytic leukemia antigen is identical to neutral endopeptidase. J Exp Med 168:1247, 1988

12. Salles G, Chen CY, Reinherz EL, Shipp MA: CDIO/NEP is expressed on Thy-l'O" B220+ murine B-cell progenitors and func- tions to regulate stromal cell dependent lymphopoiesis. Blood 80:2021, 1992

13. Greaves MF, Hariri G, Newman RA, Sutherland DR, Ritter MA, Ritz J: Selective expression of the common acute lymphoblas- tic leukemia (gp 100) antigen on immature lymphoid cells and their malignant counterparts. Blood 61:628, 1983

14. Cossman J, Neckers LM, Leonard WJ, Greene W C Poly- morphonuclear neutrophils express the common acute lymphoblas- tic leukemia antigen. J Exp Med 157:1064, 1983

15. Uckun FM, Ledbetter JA: Immunobiologic differences be- tween normal and leukemic human B-cell precursors. Proc Natl Acad Sci USA 85:8603, 1988

16. Ryan D, Kossover S, Mitchell S, Frantz C, Hennessy L, Co- hen H: Subpopulations of common acute lymphoblastic leukemia antigen-positive lymphoid cells in normal bone marrow identified by hematopoietic differentiation antigens. Blood 68:4 17, I986

17. Nadler LM, Korsmeyer SJ, Anderson KC, Boyd AW, Slau- genhoupt B, Park E, Jensen J, Coral F, Mayer RJ, Sallan SE, Ritz J, Schlossman S F B cell origin of non T-cell acute lymphoblastic leukemia. A model for discrete stages of neoplastic and normal pre-B cell differentiation. J Clin Invest 74:332, 1984

18. Hurwitz CA, Loken MR, Graham ML, Karp JE, Borowitz MJ, Pullen DJ, Civin CI: Asynchronous antigen expression in B lineage acute lymphoblastic leukemia. Blood 72:299, 1988

19. Lansdorp PM, Dougherty GJ, Humphries RK: CD34 epi- topes, in Knapp W, Dorken B, Gilks WR, Rieber EP, Schmidt RE, Stein H, Von dem Bome AEGK (eds): Leucocyte Typing IV: White Cell Differentiation Antigens. Oxford, UK, Oxford, 1990, p 826

20. LeBien TW, Wormann B, Villablanca JG, Law CL, Stein- berg LM, Shah VO, Loken MR: Multiparameter flow cytometric analysis ofhuman fetal bone marrow B cells. Leukemia 4:354, 1990

2 I . Anderson DR: Ultrastructure of normal and leukemic leu- kocytes in human peripheral blood. J Ultrastruct Res 9 5 , 1966

22. Graham RC, Karnovsky MJ: The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: Ultrastructural cytochemistry by a new technique. J Histo- chem Cytochem 14:2Y I , 1966

23. Gregory CJ, Eaves A C Human marrow cells capable of erythropoietic differentiation in vitro: Definition of three erythroid colony responses. Blood 49:855, 1977

24. Gartner S , Kaplan HS: Long-term culture of human bone marrow cells. Proc Natl Acad Sci USA 77:4756, 1980

25. Coulombel L, Eaves AC, Eaves CJ: Enzymatic treatment of long term human marrow cultures reveals the preferential location of primitive hemopoietic progenitors in the adherent layer. Blood 62:291, 1983

26. Khoury E, Lemoine FM, Baillou C, Kobari L, Deloux J, Guigon M, Najman A: Tumor necrosis factor alpha in human long- term bone marrow cultures: Distinct effects on nonadherent and adherent progenitors. Exp Hematol20:99 I , 1992

27. Andrews RG, Torok-Storb B, Bemstein ID: Myeloid asso- ciated differentiation antigens on stem cells and their progeny iden- tified by monoclonal antibodies. Blood 62: 124, 1983

28. Uckun FM: Regulation of human B-cell ontogeny. Blood 76:1908, 1990

29. Saeland S, Duvert V, Caux C, Pandrau D, Favre C, Vallk A, Durand I, Charbord P, de Vnes J, Banchereau J: Distribution of surface-membrane molecules on bone marrow and cord blood CD34' hematopoietic cells. Exp Hematol 20:24, 1992

30. McGinnes K, Letarte M, Paige CJ: B-lineage colonies from normal human bone marrow are initiated by B cells and their pro- genitors. Blood 77:96 I , 199 1

31. Whitlock CA, Witte ON: Long term culture of B lympho- cytes and their precursors from murine bone marrow. Proc Natl Acad Sci USA 79:3608, 1982

32. Lemoine FM, Humphries RK, Abraham SDM, Krystal G, Eaves CJ: Partial characterization of a novel stromal cell-derived pre-B cell growth factor active on normal and immortalized pre-B cells. Exp Hematol 16:7 18, 1988

33. Wolf ML, Buckley JA, Goldfarb A, Law CL, LeBien TW: Development of a bone marrow culture for maintenance and growth of normal human B cell precursors. J Immunol 147:3324, 1991

34. Gore SD, Kastan MB, Civin CI: Normal human bone marrow precursors that express terminal deoxynucleotidyl transfer- ase include T cell precursors and possible lymphoid stem cells. Blood 77:1681, 1991

35. Grtimayer ER, Griesinger F, Hummel DS, Brunning RD, Kersey JH: Identification of novel B-lineage cells in human fetal bone marrow that coexpress CD7. Blood 77:64, 199 1

36. Bertho JM, Mossalayi MD, Dalloul AH, Mouterde G, Debrk P Isolation of an early T cell precursor (CFU-TL) from human bone marrow. Blood 75:1064, 1990

37. Mossalayi MD, Lecron JC, Dalloul AH, Sarfati M, Bertho JM, Hofstetter H, Delespesse G, Debrk P: Soluble CD23 (FceRII) and interleukin 1 synergstically induce early human thymocyte maturation. J Exp Med 171:959, 1990

38. Mossalayi MD, Dalloul AH, Bertho JM, Lecron JC, Goude de Laforest P, Debrk P: In vitro differentiation and proliferation of purified human thymic and bone marrow CD7+ CD2- T-cell pre- cursors. Exp Hematol 18:329, 1990

39. Terstappen LWMM, Huang S, Picker LJ: Flow cytometric assessment of human T-cell differentiation in thymus and bone marrow. Blood 79:666, 1992

40. Scbmitt C, Dalloul AH, Arock M, Debrk P, Mossalayi M D In vitro evidences of T cell commitment in human bone marrow. Exp Hematol6:819, 1992 (abstr)

41. Hokland P, Rosenthal P, Griffin JD, Nadler LM, Daley J, Hokland M, Schlossman SF, Ritz J: Purification and characteriza- tion of fetal hematopoietic cells that express the common acute lymphoblastic leukemia antigen (CALLA). J Exp Med 157:114, 1983

42. Debili N, Issaad C, Masse JM, Guichard J, Katz A, Breton Gorius J, Vainchenker W: Expression of CD34 and platelet glyco- proteins during human megakaryocytic differentiation. Blood 80:3022, 1992

43. Egeland T, Steen R, Gaudernack G, Yang YC, Thorsby E: Myeloid differentiation of purified CD34+ cells after stimulation

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

STUDY OF CD34+ SUBSETS AND B-CELL PRECURSORS 429

with recombinant human granulocyte-monocyte colony-stimulat- ing factor (CSF), granulocyte-CSF, monocyte-CSF and interleukin- 3. Blood 78:3192, 1991

44. Visser JWM, de Vries P, Hogeweg-Platenburg MGC, Bayer J, Scheters G, Van Den Heuvel, Mulder DH: Culture ofhematopoi- etic stem cells purified from murine bone marrow. Semin Hematol 8:117, 1991 1992

45. Cumano A, Paige CJ, Iscove NN, Brady G: Bipotential pre- cursors of B cells and macrophages in murine fetal liver. Nature 356:612, 1992

46. Billips LG, Petitte D, Dorshkind K, Narayanan R, Chiu CP, Landreth KS: Differential roles of stromal cells, interleukin-7 and kit-ligand in the regulation of B lymphopoiesis. Blood 79: 1 185,

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom

1993 82: 417-429  

S Pontvert-Delucq, J Breton-Gorius, C Schmitt, C Baillou, J Guichard, A Najman and FM Lemoine cell subsets in relation to B-lymphoid developmentCharacterization and functional analysis of adult human bone marrow 

http://www.bloodjournal.org/content/82/2/417.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

  Copyright 2011 by The American Society of Hematology; all rights reserved.Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American

For personal use only.on March 14, 2018. by guest www.bloodjournal.orgFrom