From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
Colony Formation of Clone-Sorted Human Hematopoietic Progenitors
By Hideo Ema, Toshio Suda, Yasusada Miura, and Hiromitsu Nakauchi
To characterize human hematopoietic progenitors, w e
performed methylcellulose cultures of single cells isolated
from a population of CD34' cells by fluorescence-activated
cell-sorting (FACS) clone-sorting system. CD34' cells were
detected in bone marrow (BM) and peripheral blood (PB)
cells at incidences of 1.O% and 0.01 % of total mononuclear
cells, respectively. Single cell cultures revealed that approximately 37% of BM CD34' cells formed colonies in the
presence of phytohemagglutinin-leukocyte conditioned medium and erythropoietin. Erythroid bursts-, granulocytemacrophage (GM) colony-, and pure macrophage (Mac)
colony-forming cells were 10%each in CD34' cells. Approximately 15% of PB CD34+ cells formed colonies in which
erythroid bursts were predominant. CD34' cells were
heterogeneous and fractionated by several antibodies in
FACS multicolor analysis. In these fractionated cells, CD34+,
CD33' cells formed GM and Mac colonies 7 t o 10 times as
often as CD34'. CD33- cells. Most of the erythroid bursts
and colonies were observed in the fraction of CD34'.
CD13- cells or CD34+, CD33- cells. The expression of
HLA-DR on CD34' cells was not related t o the incidence,
size, or type of colonies. There was no difference in the
phenotypical heterogeneity of CD34' cells between BM
and PB. About 10% of CD34' cells were able t o form G
colonies in response t o granulocyte colony-stimulating
factor (G-CSF) and t o form Mac colonies in GM-CSF or
interleukin-3 (IL-3). Progenitors capable of generating colonies by stimulation of G-CSF were more enriched in CD34'.
CD33' fraction than in CD34'. CD33- fraction. Thus,
single cell cultures using the FACS clone-sorting system
provide an accurate estimation of hematopoietic progenitors and an assay system for direct action of colonystimulating factors.
0 1990by The American Society of Hematology.
E
informed consent. These samples were diluted with phosphatebuffered saline (PBS), layered over Ficoll-Metrizoate (Lymphoprep;
Nyegaad, Oslo, Norway), and centrifuged for 30 mintues at 400gat
room temperature. Interface cells were washed twice with PBS,
passed through a stainless mesh, and washed once more with staining
medium (3% fetal calf serum (FCS) and 0.1% sodium azide in PBS).
BM and PB mononuclear cells were pelleted for staining with
monoclonal antibodies (MoAbs).
Monoclonal antibodies. HPCA- 1 (IgGl), avidin phycoerythin
(PE)-conjugated Leu-4 (IgGl), biotinylated Leu-3a (IgGl), PEconjugated Leu-1 (IgGZa), biotinylated Leu-2a (IgGl), PEconjugated Leu-1 5 (IgG2a), PE-conjugated Leu-M3 (IgG2a), LeuM1 (IgM), PE-conjugated Leu-12 (IgGl), and PE-conjugated
HLA-DR (IgG2a, reacts with a nonpolymorphic HLA-DR epitope)
were purchased from Becton-Dickinson Immunocytometry Systems,
Inc, Mountain View, CA. PE-conjugated J5 (IgG2a), PE-conjugated My7 (IgGl), and PE-conjugated My9 (IgG2a) were purchased from Coulter Immunology, Hialeah, FL. Anti-human glycoprotein (GP) IIb/IIIa complex (IgM) and OKT16 (IgG2a) were
purchased from Dakopatts, Glostrup, Denmark and Ortho Diagnostic Systems, Inc, Raritan, NJ. Fluorescein isothiocyanate (FITC)conjugated rat anti-mouse IgGl (Zymed Laboratories, Inc, San
Francisco, CA), biotinylated rat anti-mouse IgG2a (Zymed Laboratories), and biotinylated rat anti-mouse p (PharMingen, San Diego,
CA) were used as second antibodies for unconjugated antibodies:
HPCA-1, OKT16, Leu-M1, and anti-GP IIb/IIIa. Biotinylated
N R I C H M E N T OF human hematopoietic progenitors
has been done by positive selection using monoclonal
antib~dies.'.~
CD34 antigen is expressed on colony-forming
cells that consist of erythroid, granulocyte-macrophage, and
megakarocyte colony-forming units. It has been reported
that CD34 positive (CD34+) cells constitute approximately
1% to 4% of normal human bone marrow cells, and they are
not detectable in normal human peripheral
By using
the technique of complement dependent cytot~xicity,~
several monoclonal antibodies have been shown to react with
hematopoietic progenitors. However, those antibodies also
reacted with a large number of more mature cells other than
hematopoietic progenitors.6 CD34+ cells still have heterogeneous surface antigens. Anti-CD13,' CD33,8 and HLA-DR
monoclonal antibodies react with a fraction of CD34' cells.'
Andrews et a1 demonstrated that colony-forming cells were
derived from CD33-, CD34+ cells, rather than CD33+,
CD34+ cells, in long-term marrow cultures, and the latter
contained more colony forming cells than the former before
culturing." Fractionated CD34+ cells have been characterized by the expression of HLA-DR. Colony-forming cells
were most enriched in the fraction of cells expressing high
CD34 and low HLA-DR." Blast cell colony forming units
and burst-forming units-magakaryocyte expressed CD34,
but not HLA-DR.I2,I3
To clarify the colony-forming capacity of single cells and
the direct effect of purified hematopoietic factors on target
cells without the influence of other surrounding cells, single
cell cultures are required. With the development of the
clone-sorting system of fluorescence-activated cell sorting
(FACS),I4 it became possible to deposit single cells in
semisolid culture medium.I5*l6
Therefore, we studied the colony-forming capacity of
single bone marrow and peripheral CD34+ cells, determined
the phenotypes of colony-forming cells and individual colonyforming units, and evaluated the incidence of hematopoietic
factor-responsive cells using the FACS clone-sorting system.
MATERIALS AND METHODS
Cells. Bone marrow (BM) and peripheral blood (PB) samples
were obtained from normal healthy volunteers who had given
Blood, Vol 75,No 10 (May 15). 1990: pp 1941-1946
From the Division of Hematology, Department of Medicine, Jichi
Medical School, Tochigi-ken; and the Laboratory of Molecular
Regulation of Aging, Frontier Research Program, The Institute of
Physical and Chemical Research (RIKEN).Ibaraki-ken, Japan.
Submitted November 11.1989; accepted January 17,1990.
Supported in part by Grants-in-Aid from the Ministry of Education, Science and Culture of Japan.
Address reprint requests to Hideo Ema, MD. Division of Hematology. Department of Medicine. Jichi Medical School, Minamikawachi-machi, Tochigi-ken, 329-04, Japan.
The publication costs of this 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 io
indicate this fact.
0 1990 by The American Society of Hematology.
0006-4971/90/7510-0015$3.00/0
1941
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1942
EMA ET AL
antibodies were labeled with streptavidin-PE (SAV-PE) or streptavidin-allophycocyanin (SAV-APC).
Staining procedures. For double staining, cells were incubated
with HPCA-1 and one of the different subclass antibodies, either
biotinylated or directly PE-conjugated for 30 minutes on ice, washed
twice, and treated with anti-mouse IgGI-FITC for 30 minutes on
ice. SAV-PE or SAV-APC was used when another antibody was
biotinylated. When the isotype of another antibody was also IgG1,
another antibody was added after incubation with 20% normal
mouse serum to block the remaining free sites of anti-IgG1-FITC.
Cells were suspended in staining medium containing 7-aminoactinomycin D (7-AAMD). For negative controls, unstained cells or cells
stained with only second antibodies were included.
Analysis by FACS. Stained cells were analyzed on FACStar
plus (Becton-Dickinson). Multiparameter data were collected and
analyzed using FACS-DESK (version 1.2) run on a Digital Micro
VAX-I1 GPX, configured as described.” Fluorescence intensity of
individual cells was measured as relative fluorescence units. Data
from more than 50,000 cells for bone marrow samples and more than
100,000 cells for peripheral samples were collected and analyzed.
Establishment of the gate was based on the light scattering properties of CD34+ cells in each experiment.
Clone-sorting. A sort gate was set for CD34 bright cells (channel of FITC greater than about 110) in each experiment in order to
eliminate contamination by CD34- cells, as a difference in colonyforming ability was not observed between CD34 bright cells and
CD34 dull cells. CD34+ cells and fractionated CD34+ cells by the
expression of other antigens were sorted. FACStar plus and automatic cell deposition unit (Becton-Dickinson) were used for single
cell deposition into each well of 96-well plates. The dead cells stained
with 7-AAMD were gated out by FACS at the time of cell sorting.
Single cell cultures. Methylcellulose cultures were done using
conditions suitable for the growth of all myeloid lineages, including
megakaryocyte colony formation, as previously described.” Briefly,
0.1 mL of culture medium was placed in each well of a 96-well flat
bottom microtiter plate (A/S Nunc, Kampstrup, Denmark) before
the clone-sorting. Culture medium contained 1.2% methylcellulose
(Aldrich Chemical Co, Milwaukee, WI), 30% human platelet-poor
plasma (PPP), 1% deionized bovine serum albumin (BSA, Sigma
mol/L 2-mercaptoethanol
Chemical Co, St Louis, MO), 5 x
(2-ME Sigma), 5% phytohemagglutinin-stimulated leukocyte conditioned medium (PHA-LCM),I9 and 2 U/mL recombinant human
erythropoietin (Epo; provided by Snow Brand Co, Ishibashi, Japan)
in Iscove’s modified Dulbecco‘s medium (IMDM; GIBCO, Grand
Island, NY). To examine the direct effects of hematopoietic factors
on a single CD34’ cell, 20 ng/mL recombinant human granulocyte
colony-stimulating factor (G-CSF; specific activity, 2.5 x 1O7 U/mg
protein; provided by Chugai Pharmaceutical Co, Tokyo, Japan), 200
U/mL recombinant human granulocyte-macrophage colony-stimulating factor (GM-CSF; specific activity, 1.0 x lo9 U/mg protein,
provided by Sumitomo Pharmaceutical Co, Osaka, Japan), 100
U/mL recombinant human interleukin-3 (IL-3; specific activity;
21,900 U/mL, provided by Genetics Institute, Cambridge, MA), 1
ng/mL recombinant human interleukin-5 (IL-5; purity of 80%,
provided by Suntory Central Research Institute, Osaka, Japan), or
20 ng/mL recombinant human interleukin-6 (IL-6; specific activity,
3.9 x lo9 U/mg protein, provided by Ajinomoto Pharmaceutical
Co, Kawasaki, Japan) was added to culture medium. In these
cultures, 30% PPP was replaced by 30% FCS (Flow Laboratories,
North Ryde, NSW, Australia). The cultures were incubated for 15
days at 37OC in a humidified atmosphere of 5% CO, and 5% 0,. On
days 10 and 15 of culture, colony and cluster formation were
observed and classified using an inverted microscope. To confirm
colony types, each colony was lifted from the semisolid medium on
day 15, and Cytospin (Shandon Southern Instruments Inc, Sewickley, PA) preparations were made and stained with May-GrunwaldGiemsa for morphologic analysis of colony cells. As the colony
formation in the presence of IL-3 was delayed, the observation
period was prolonged to 20 or more days. Colonies and clusters were
defined as aggregates consisting of 40 or more cells and 8 to 39 cells,
respectively. Colony types were also defined as follows: erythroid
bursts are colonies consisting of erythroblasts detected on day 15 of
culture; erythroid (CFU-E-derived) colonies consist of erythroblasts
detected before day 10; granulocyte/granulocyte-macrophage(G/
GM) colonies consist of only granulocytes or granulocytes and
macrophages (monocytes); pure macrophage (Mac) colonies consist
of only macrophages and/or monocytes; megakaryocyte (Meg)
colonies consist of three or more megakaryocytes; and mixed (Mix)
colonies consist of both erythroblasts and granulocytes/macrophages.
RESULTS
CD34’ cells analyzed and sorted by FACS. To obtain
enriched hematopoietic progenitors, mononuclear cells either
from BM or PB were stained with anti-CD34 (HPCA-1) and
FITC-conjugated rat anti-mouse IgG 1 antibodies. These
cells were then subjected to FACS analysis and clonesorting. Figure 1 shows the light scattering properties of
CD34+ cells. CD34+ cells had a relatively narrow range of
side scattering (SSC) and a relatively wide range of forward
scattering (FSC). We set the gate using these ranges of SSC
and FSC. Most of the granulocytes were excluded from the
gate, and CD34’ cells were easily detected by setting the
gate. BM and PB mononuclear cells within the gate accounted for 36.2% 2 11.0% (mean 2 SD) and 52.2% f
12.7% of the total mononuclear cells, respectively. Bone
marrow CD34 bright cells used in our study were 1.05% 2
0.44% of the gated cells and 0.40% * 0.23% of the total
mononuclear cells. Both CD34 bright and dull cells accounted for about 1% of the total mononuclear cells. PB
w-
M
n
0
ssc
FSC
Fig 1. SSC and FSC gates for CD34’ cells.
Channels of SSC were 85 to 105 those of FSC. 90
to 180 (expressed as mean values of 12 samples).
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
CLONE-SORTED HUMAN HEMATOPOIETIC PROGENITORS
1943
Table 1. Colony Formation of Single BM and PE CD34' Calls in the Presence of PHA-LCM and Epo
Incidence of Cdony-Forming Cells (%)
Clone-Sated Cells
BM CD34+
PB CD34'
BFU-E
CFU-E
GIGM
Mac
Mix
Mea
Total
10.0 f 5.7
7.5 i 1.7
5.1 i 8.3
0
10.2 f 3.3
3.8 f 4.1
10.7 f 5.7
2.9 f 1.5
1.0 f 1.0
0.4 f 0.0
0
0
37.1 f 14.0
14.6 f 1.6
~
~~
Data of 720 clonesorted BM CD34+ cells and 240 clone-sorted PB CD34' cells were analyzed.
CD34+ cells accounted for 0.021% k 0.007% of the gated
cells and 0.01 1% k 0.002% of the total mononuclear cells. It
was difficult to separate CD34 dull cells from CD34 bright
cells in PB.
Colony formation by single CD34+ cells in the bone
marrow and peripheral blood. We evaluated a clonesorted cell capable of proliferating and differentiating in the
methylcellulose cultures for 15 to 20 days. Table 1 summarizes the results of the colony formation by BM and PB
CD34+ cells in the presence of PHA-LCM and Epo. These
data were obtained from the analysis of 12 bone marrow
samples (total, 720 clone-sorted cells) and four peripheral
samples (total, 240 clone-sorted cells). An accuracy of single
cell-deposition was tested by depositing one bead in one well
3,000 times. Errors were identified in only 15 wells of them.
We never observed two or more colonies in a single well of a
microtiter plate. Colony-forming cells were markedly enriched in this population. They represented 37.1% * 14.0%
(mean * SD) of CD34+ cells in the bone marrow, and only
0.02% of the pooled CD34- cells. Incidence of erythroid
bursts, G/GM colonies, and Mac colonies were similarly
around 1 / 10 of CD34+ cells. Incidence of mixed colonies was
about 1/100 of CD34+ cells. Meg colonies could not be
detected in the single cell experiments, but they were
observed in cultures of pooled CD34' cells or unseparated
mononuclear cells.
On the other hand, colony-forming cells accounted for
14.6% 1.6% (mean k SD) of peripheral CD34+ cells. The
incidence of erythroid bursts was higher than that of other
Fig 2.
Representative anal-
yses of bone marrow and peripheral blood stained with
HPCA-1 (anti-CD34) labeled by
anti-lgG1-FITC and My7-PE
(anti-CD13)(A and D); HPCA-1
and MyS-PE (anti-CD33) (B and
E); and HPCA-1 and HLA-DRPE (C and F). The distribution of
bone marrow CD34+ cells (A,
B. and C) resembles that of
peripheral CD34' cells (D, E,
and F). Bone marrow cells are
represented by 1% probability
contour plotting; peripheral
blood cells, by 50% logarithmic
contour spacing. Peripheral
CD34' cells were not detected
at all in the usual 1% probability
contour plotting. but 50% logarithmic contour spacing made
them visible on a contour map.
The sorting window of each
fraction is indicated in these
maps.
colony types in the peripheral blood. Erythroid colonies were
detectable in BM CD34+ cells, but not in PB CD34+ cells.
Subpopulations of CD34+ cells and colony formation of
clone-sorted cells. CD34+ cells were further stained with
anti-CD3, CD4, CD5, CD7, CD8, CDIO, CDll, CD13,
CD14, CD15, CD19, CD33, CD41, and HLA-DR. The
representative FACS analyses are shown in Fig 2. The
proportions of double positive cells to the total CD34+ cells in
the bone marrow correlated with those in the peripheral
blood, though there were variations in these antigens' coexpressions with CD34. Bone marrow CD34+ cells were
negative for anti-CD5, CD7, CD8, CD11, CD14, CD15, and
CD41. Percentages of positive cells for anti-CD3, CD4, and
CD19 were markedly varied among individuals. For example, CD4 was expressed in 1.O% to 21.7% of the total CD34+
cells from four donors.
The number of colonies was similar between positive and
negative cells for anti-CD13 and HLA-DR (Table 2). The
CD33 bright subpopulation contained twice as many colonyforming cells as the CD33 negative subpopulation of CD34+
cells. The gates set for clone-sorting of these cells were
indicated in Fig 2. There were differences in the types of
colonies between positive and negative fractions for antiCD13 and CD33. Figure 3 shows the changes of colony types
by antigen expression. The number of each type of colony in
the positive or negative fractions was divided by that of the
corresponding colony type in the total CD34+ cells, and the
results were expressed as ratios. Values larger than 1.00
meant that a larger number of colonies was formed by the
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
EMA ET AL
1944
Table 2. Subpopulations of BM CD34' Cells and Colony
Formation of Clone-Sorted Cells
Colony Formation by
CD
% Positive Cells
in BM 0 3 4 ' Cells
MoAb
Positive
Fraction
Negative
Fraction
1/120
491120
21120
361120
431120
401120
511120
371120
01120
561180
261120
-
*
CDlO
J5
CD13
CD15
My7
Leu-M1
CD33
My9
Bright
Dull
GPllb/llla
HLA-DR
CD41
-
12.5 4.6
38.8
36.1
1.6 t 2.2
54.4 23.3.
11.7 2 5.5
47.2 k 21.6
2.2 f 0.3
83.4 2 7.3
*
*
481120
521180
Percentages expressed as mean f SD (n z 3) the number of colonies
per 120 to 180 clone-sorted cells.
"Bright plus dull positive cells.
subpopulation compared with the positive control. The incidences of erythroid bursts- and eythroid colony-forming cells
in CD34+, CD13- cells and CD34+, CD33- cells were
significantly higher than those in CD34+, CD13+ cells and
CD34+, CD33+ cells. The subpopulation of CD34+, CD33
bright cells formed 7 to 10 times as many G/GM and Mac
colonies as that of CD34+, CD33- cells. The incidence, type,
and size of the colonies were not dependent on the expression
of HLA-DR on CD34+ cells. Erythroid (CFU-E-derived)
CD13
2.5
2
colonies were observed only in subpopulations of CD 13-,
CD33- and HLA-DR+ cells within gated CD34+ cells.
Direct effects of hematopoietic factors on single CD34+
cells. In order to study the effects of hematopoietic factors
on myeloid colony formation by single progenitors, clonesorted BM CD34+ cells were cultured in the presence of
G-CSF, GM-CSF, IL-3, IL-5, or IL-6 (Table 3). About 10%
of the CD34+ cells generated G colonies in the presence of
G-CSF or Mac colonies in the presence of GM-CSF or IL-3.
The incidence of G colony formation in the presence of
G-CSF was similar to that in the presence of PHA-LCM.
Mac colonies were predominantly formed in the presence of
GM-CSF or IL-3. The sizes of G/GM colonies induced by
GM-CSF or IL-3 were smaller than those induced by
G-CSF. About 85% of the total GM colonies in the presence
of IL-3 were formed from day 15 to day 20. These colonies
consisted of 40 to 50 cells that were morphologically immature granulocytes and monocytes. Some of the cells generated secondary macrophage clusters in the presence of
PHA-LCM plus Epo after replating (data not shown). No
eosinophil or G/GM colonies were formed by single CD34+
cells in the presence of IL-5 or IL-6. Effects of G-CSF or
IL-3 or, fractionated cells by CD33 expression were also
examined. G-CSF responsive colony-forming cells were detected in CD34+,CD33+ fraction, but not in CD34+,CD33-
A
n
B
CD33
3.51
3l
s2
2.5
01.5
.-*
(0
=
(R
a1.5
1
1
0.5
0.5
0
Total
CFCs
BFU-e
G/GM
Mac
CFU-e
n
-
Total
BFU-e
G/GM
Mac
CFU-e
CFCs
Types of colonies
Types of colonies
C
HLA-DR
2.5
c
0 1.E
._
c
(D
a
1
0.5
0
d
I
Tota I
BFU-e
CFCs
G/GM
Mac
Types of colonies
1_
CFU-e
., ., .,
Fig 3. Colony formations of negative (-1 and positive ( + I
fractions for antLCD13 (A:
CD13-; 0. CD13'). CD33 (B:
CD33-; 0 , CD33+'-; B, CD33"). and HLA-DR (C:
HLA-DR-; 0.
HLA-DR+) antibodies in CD34' cells sorted by FACS. Ratio indicates
the number of one of the colony types formed by the positive or
negative fraction divided by the number of the same type of colonies
formed by the total CD34' cells. These columns expressed only
mean ratios of three experiments where 120 to 180 clone-sorted
cells were cultured. + indicates bright; / -, dull positive cells.
+
+
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
CLONE-SORTED HUMAN HEMATOPOIETIC PROGENITORS
Table 3. Colony Formation of Single BM CD34' Cells in the
Presence of Each Hematopoietic Factor
No. Colonies
Cells/Factor
CD34'
None
G-CSF
GM-CSF
IL-3
IL-5
IL-6
CD34'. CD33'
G-CSF
IL-3
CD34', CD33G-CSF
IL-3
GIGM
0
37
6
25
Mac
0
81360
181360
391360
361360
41360
241360
18
8
71 120
161120
1
01120
11120
ot
11
Numbers of colonies formed by 3 6 0 clone-sorted CD34' cells and
120 clonesorted CD34'. CD33' or CD34'. 0 3 3 - cells. CD34' cells
were divided into CD33' and CD33- fractions, which were different
from the sorting windows shown in Fig 2 (CD34'. CD33' cells partially
included CD34'. CD33"- cells).
*Of 2 5 GM colonies, 2 1 consisted of only immature granulocytes and
monocytes.
tNeither GM colonies nor eosinophil colonies were formed in the
presence of IL-5.
fraction. However, IL-3 responsive colony-forming cells
existed in both fractions. The sorting window of CD34+,
CD33' cells was different from that shown in Fig 2, and
partially included CD34+, CD33+/- cells in this experiment.
DISCUSSION
The culture system for single cells is useful for detecting
individual hematopoietic progenitors capable of generating
colonies and for examining the direct effects of purified
hematopoietic factors on hematopoietic progenitors.l6.*' We
used the FACS clone-sorting system for culturing single cells
sorted from a CD34+ cell population containing hematopoietic progenitors in bone marrow and peripheral blood cells.
This system made it possible to evaluate the colony-forming
capacity of single hematopoietic progenitors and to examine
the direct action of growth factors on them, without the
influences of secreted cytokines or cell-cell interactions with
the cells co-existing in culture. Furthermore, a very small
fraction of cells, such as CD34+ cells in the peripheral blood,
can be analyzed precisely, and only several hundred cells are
required for single cell experiments.
We set the gate based on the light scattering properties of
CD34+ cells for the accurate sorting of CD34+ cells. CD34+
cells had a relatively narrow range of SSC and a relatively
wide range of FSC. This finding indicated that the sizes of
CD34+ cells were varied. Andrews et all0 and Sutherland et
aIz' studied CD34+ cells by separating these cells based on
cell sizes. CD34' cells of all sizes were used as target cells in
our study.
Colony-forming cells were highly enriched in the population of CD34+ cells and were few in the population of CD34cells. CD34+ cells were heterogeneous, and contained different kinds of colony forming cells. We compared bone
marrow CD34+ cells with peripheral CD34+ cells in respect
1945
to colony-forming ability. The former contained nearly
2.5-fold more colony-forming cells than the latter. Total
colonies and clusters accounted for 45% and 18% of the total
CD34+ cells in the bone marrow and peripheral blood,
respectively. The proportion of erythroid bursts G/GM
co1onies:Mac colonies in bone marrow CD34+ cells was
approximately 1:l:l and was different from that in peripheral CD34+ cells, where erythroid progenitors predominated.
However, obvious differences in the phenotypes of CD34+
cells were not recognized between the bone marrow and
peripheral blood by means of the two-color analysis of
CD34+ cells with anti-CD13, CD33, HLA-DR, and other
MoAbs. Civin et a13demonstrated that bone marrow CD34+
cells were negative for anti-CD11, CD14, CD15, and CD16.
Nagler et alZ2reported that peripheral CD34+ cells, which
were enriched by the negative selections, did not coexpress
mature lymphocyte antigens, such as CD2, CD3, CD4, CD5,
and CD8. Their findings are consistent with our data. CD3
and CD4 were each detected in about 10% of CD34+ cells in
some samples. We could not negate nonspecific bindings,
though the cells expressing CD3 or CD4 with CD34 did not
form any myeloid colonies.
As Griffin et als suggested that anti-CD33 is useful for the
purification of subsets of colony-forming cells, we were able
to select G/GM and Mac colony-forming cells from CD34+
cells by CD33 expression. However, CD13 obviously did not
recognize unipotential progenitors. Most of the erythroid
progenitors (BFU-E and CFU-E) in CD34+ cells did not
coexpress CD13 or CD33. CD34+, HLA-DR- cells generated colonies with similar efficiency to CD34+, HLA-DR+
cells, though they were much smaller in population. The
expression of HLA-DR did not affect unipotential colony
formation by CD34+ cells. HLA-DR is expressed in a
majority of BFU-E, but is progressively lost during differentiation through CFU-E to erythroblast^.^^ This is not contradictory to our data. The incidence of BFU-E was not
different between CD34+, HLA-DR+ cells and CD34+,
HLA-DR- cells, but HLA-DR+ cells comprised a large
proportion of CD34+ cells. We assumed that CFU-E in
CD34+ cells are at differentiating stages close to BFU-E.
Recently, Caux et alZ4demonstrated that CD34+, HLADR+ cord blood cells cultured in the presence of IL-3 give
rise to a CD34-, HLA-DR+ population with a high proliferative capacity, as determined by 3H-TdR uptake. HLA-DR
might also be related to differentation at the later stages.
Meg colonies were not observed in single cell experiments
using CD34' cells, which may be due to a very low incidence
of megakaryocyte progenitors in CD34+ cells or inadequate
conditions to Meg colony formation by single progenitors.
Single cell cultures clearly demonstrated the direct effects
of hematopoietic factors on myeloid colony formation by
CD34+ cells. G-CSF supported G colony formation in 10%of
the single CD34+ cells; however, GM-CSF supported colony
formation in only 1% to 2% of them. As reported
previously,25s26
GM-CSF might require cell-cell interactions
or synergistic factors for full stimulation of GM colony
formation. Most of the GM colonies formed in the presence
of IL-3 consisted of only immature granulocytes and monocytes. Thus, IL-3 did not always support the terminal
differentiation of granulocytes like G-CSF. In the presence
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1946
EMA ET AL
of both G-CSF and IL-3, the size of each G/GM colony
became large, but the number of G/GM colonies did not
increase compared with the presence of only G-CSF (data
not shown). G-CSF alone might stimulate all G/GM progenitors in CD34+ cells. Furthermore, G-CSF acted more
effectively to form colonies on CD34+, CD33+ cells than on
CD34+, CD33- cells. IL-5 did not support GM or eosiniphil
colony formation of single CD34+ cells. Eosinophil progenitors as target cells for IL-5 were more mature cells than the
CD34+ c e k 2 ’
It has been difficult to distinguish self-renewing, pluripo-
tent hematopoietic stem cells from multipotent, nonrenewing, hematopoietic
Most of the total CD34+
cells capable of generating colonies were unipotential progenitors as described above, although mixed colonies and megakaryocyte colonies were formed from CD34’ cells.
ACKNOWLEDGMENT
We thank K. Nagayoshi for FACS operation, S. Kurokawa
for skillful technical assistance, and M. Yoshida for preparing the manuscript.
REFERENCES
1. Civin IC, Strauss LC, Brovall C, Fackler MJ, Schwartz JF,
Shaper JH: Antigenic analysis of hematopoiesis. 111. A hematopoietic progenitor cell surface antigen defined by a monoclonal antibody
raised against KG-la cells. J Immunol 133:157, 1984
2. Strauss LC, Rowley SD, La Russa VF, Sharkis SJ, Stuart RK,
Civin CI: Antigenic analysis of hematopoiesis. V. Characterization
of My- 10 antigen expression by normal lymphohematopoieticprogenitor cells. Exp Hematol 14878, 1986
3. Civin CI, Banquerigo ML, Strauss LC, Loken M R Antigenic
analysis of hematopoiesis. VI. Flow cytometric characterization of
My-IO-positive progenitor cells in normal human bone marrow. Exp
Hematol 15:10, 1987
4. Bodger M P Isolation of hemopoietic progenitor cells from
human umbilical cord blood. Exp Hematol 155369, 1987
5. Reinherz EL, Haynes BF, Nadler LM, Bernstein ID (eds):
Leukocyte Typing 11, vol 2. New York, NY, Springer-Verlag. 1986
6. Andrews RG, Torok-Storb B, Bernstein ID: Myeloid-associated differentiation antigens on stem cells and their progeny identified by monoclonal antibodies. Blood 62124,1983
7. Griffin JD, Ritz J, Beveridge RP, Lipton JM, Daley JF,
Schlossman SF Expression of MY7 antigen on myeloid precursor
cells. Int J Cell Cloning 1:33,1983
8. Griffin JD, Linch D, Sabbath K, Larcom P, Schlossman SF: A
monoclonal antibody reactive with normal and leukemic myeloid
progenitor cells. Leuk Res 8521, 1984
9. Folks TM, Kessler SW, Orenstein JM, Justement JS, Jaffe ES,
Fauci AS: Infection and replication of HIV-1 in purified progenitor
cells of human bone marrow. Science 242:919, 1988
10. Andrews RG, Singer JW, Bernstein ID: Precursors of colonyforming cells in humans can be distinguished from colony-forming
cells by expression of the CD33 and CD34 antigens and light scatter
properties. J Exp Med 169:1721, 1989
11. Lu L, Walker D, Broxmeyer HE, Hoffman R, Hu W, Walker
E Characterization of adult human marrow hematopoietic progenitors highly enriched by two-color cell sorting with My10 and major
histocompatibility class I1 monoclonal antibodies. J Immunol 139:
1823,1987
12. Brandt J, Baird N, Lu L, Srour E, Hoffman R Characterization of a human hemopoietic progenitor cell capable of forming blast
cell containing colonies in vitro. J Clin Invest 82:1017, 1988
13. Briddell RA, Brandt JE, Straneva JE, Srour EF, Hoffman R:
Characterization of the human burst-forming unit-megakaryocyte.
Blood 74:145, 1989
14. Parkes DR, Bryan VM, Oi VT, Herzenberg LA: Antigenspecific identification and cloning hybridomas with a fluorescenceactivated cell sorter. Proc Natl Acad Sci USA 76:1962, 1979
15. Kannourakis G , Johnson GR, Battye F: Clonal proliferation
in vitro of individual murine and human hemopoietic cells after
fluorescence-activated cell sorting. Exp Hematol 16:367, 1988
16. Suda T, Ohara A, Suda J, Okada S, Tokuyama N, Miura Y,
Sudo T, Nishikawa S, Nakauchi H: Early B cell differentiation from
hematopoietic stem cells in the presence of stromal cells and
interleukin-7 (IL-7). Exp Hematol Today (in press)
17. Moore W, Kantz R Data analysis in flow cytometry, in Weir
DM, Herzenberg LA, Blackwell C, Herzenberg LA (eds): The
Handbook of Experimental Immunology, 4th ed. Edinburgh, Scotland, Blackwell, 1986, p 301
18. Messner HA, Jamal N, Izaguirre C: The growth of large
megakaryocyte colonies from human bone marrow. J Cell Physiol
1:45,1982 (suppl)
19. Kimura H, Burstein AS, Thorning D, Powell JS, Harker LA,
Fialkow PI, Adamson JW: Human megakaryocytic progenitors
(CFU-M) assayed in methylcellulose: Physical characteristics and
requirements for growth. J Cell Physiol 118:87, 1984
20. Metcalf D: The Molecular Control of Blood Cells. Cambridge, MA, Harvard University Press, 1988
21. 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 741563,1989
22. Nagler A, Greenberg PL, Lanier LL, Phillips JH: The effects
of recombinant interleukin 2-activated natural killer cells on autologous peripheral blood hematopoietic progenitors. J Exp Med 168:47,
1988
23. Robinson J, Sieff C, Delia D, Edwards PAW, Greaves M:
Expression of cell surface HLA-DR, HLA-ABC and glycophorin
during erythroid differentiation. Nature 289:68,1981
24. Caux C, Favre C, Saeland S, Duvert V, Mannoni P, Durand I,
Aubry J-P, de Vries JE: Sequential loss of CD34 and class I1 MHC
antigens on purified cord blood hematopoietic progenitors cultured
with IL-3: Characterization of CD34-, HLA-DR+ cells. Blood
74:1287,1989
25. Ferrero D, Tarella C, Badoni R, Caracciolo D, Bellone G,
Pileri A, Gallo E: Granulocyte-macrophage colony-stimulating factor requires interaction with accessory cells or granulocyte-colony
stimulating factor for full stimulation of human myeloid progenitors.
Blood 73:402,1989
26. Namiki M, Hara H: Enhancement of colony-formingactivity
of granulocyte-macrophage colony-stimulating factor by monocytes
in vitro. Blood 74:918, 1989
27. Ema H, Suda T, Miura Y, Nakauchi H: Manuscript in
preparation
28. Spangrude GJ, Heimfeld S, Weissman IL: Purification and
characterization of mouse hematopoietic stem cells. Science 241 5 8 ,
1988
29. Szilvassy SJ, Lansdorp PM, Humphries K, Eaves AC, Eaves
CJ: Isolation in a single step of highly enriched murine hematopoietic stem cell population with competitive long-term repopulating
ability. Blood 74:930, 1989
From www.bloodjournal.org by guest on June 15, 2017. For personal use only.
1990 75: 1941-1946
Colony formation of clone-sorted human hematopoietic progenitors
H Ema, T Suda, Y Miura and H Nakauchi
Updated information and services can be found at:
http://www.bloodjournal.org/content/75/10/1941.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.