

* Paterson Institute for Cancer Research, Manchester, United Kingdom; and
Kanazawa University, Ishikawa, and
Molecular Preventive Medicine, University of Tokyo, Japan
Correspondence: E. A. de Wynter, Molecular Medicine Unit, Clinical Sciences Building, St Jamess University Hospital, Leeds LS9 7TF, United Kingdom. E-mail: medeadw{at}leeds.ac.uk
|
|
|---|
Key Words: myeloid progenitors CCR1 antibody CD34+ cells MIP-1
|
|
|---|
Macrophage inflammatory protein-1
(MIP-1
), a cysteine
cysteine (CC) chemokine, has diverse actions that inhibit or enhance
proliferation of subpopulations of myeloid progenitor cells
[12
13
14
15
], inhibit stem cell proliferation
[16
17
18
19
20
], and mobilize progenitor cells into the blood
[16
, 21
]. The enhancing effect of MIP-1
on colony formation by granulocyte-macrophage colony-forming cells
(GM-CFCs) was more apparent on progenitor cells isolated from cord
blood [14
]. Under similar assay conditions, growth of
GM-CFCs from bone marrow cells was suppressed. The effects of MIP-1
on hemopoietic progenitors are direct and appear to be mediated by
specific receptors [15
]. At least three chemokine
receptors, CCR1, CCR5, and D6, bind to MIP-1
, although D6 does not
signal in response to interaction with the molecule
[22
23
24
25
]. It is not clear which of these receptors
mediate the different MIP-1
responses.
mRNA for CCR1 is expressed abundantly in CD34+ cells isolated from bone marrow and cord blood [14 ]. Recently, a specific antibody to the CCR1 receptor has been described, and the receptor has been shown to be expressed on some lymphocytes and monocytes but not on granulocytes [26 ]. Here we show that the majority of CD34+ cells expressed high levels of CCR1 and that this expression may be used to separate erythroid and granulocyte/macrophage progenitors. Therefore, expression of the CCR1 receptor on CD34+ cells allows rapid discrimination between erythroid and granulocyte/macrophage progenitors.
|
|
|---|
BB10010 (British Biotech, Oxford, U.K.) BB10010 is an active,
nonaggregating variant of human MIP-1
that exhibits similar
biological effects to native MIP-1
[27
,
28
].
Isolation and enrichment of CD34+ cord blood cells
Human cord blood was obtained from normal, full-term
deliveries with the local ethical committees approval. Samples were
diluted 1:1 with phosphate buffered saline (PBS) and centrifuged over
Ficoll-Hypaque gradients (1.077 g/mL) (Lymphoprep; Life Technologies)
for 25 min at 400 g. Mononuclear cells (MNCs) were recovered
from the interface, washed in PBS, and resuspended in ice-cold PBS
containing 5 mM EDTA and 0.5% (w/v) bovine serum albumin (PBE).
CD34+ cells were isolated as previously described [14 ] using the Miltenyi Isolation System (Miltenyi Biotech, Bergisch Gladbach, Germany). Briefly, MNCs were labeled with CD34 antibody conjugated to superparamagnetic beads by incubating them for 30 min at 4°C. The cells were washed, and the CD34+ cells were separated using MiniMACS columns according to the manufacturers instructions. The CD34-enriched fractions were collected by flushing the column with cold PBE buffer.
Labeling of MNCs and CD34+ cells with CCR1 antibody
To analyze CCR1 expression, 510 x 105 MNCs
or 25 x 105 CD34+ cells were first
incubated in antibody diluent (DAKO Ltd., Ely, U.K.) containing 1%
(v/v) human AB serum and previously titrated anti-CCR1 antibody at an
optimal concentration of 25 µg/mL [26
]. After the
cells were washed, primary antibody was detected with a secondary swine
anti-rabbit fluorescein isothiocyanate (FITC)-conjugated antibody (DAKO
Ltd.). For two-color analysis, the washed CD34+ cells were
labeled first with anti-CCR1 and then with swine anti-rabbit-FITC. They
were subsequently labeled with anti-CD34 phycoerythrin (PE)-conjugated
antibody (HPCA-2) (Becton Dickinson, Cowley, Oxford, U.K.). Portions of
the cells were stained with the appropriate control antibodies. All
labeled cells were analyzed and sorted by flow cytometry.
Fluorescein-activated cell sorter analysis and cell sorting
Analysis of the labeled cells was performed with a FACS Vantage
flow cytometer (Becton Dickinson, San Jose, CA) equipped with a 488-nm
argon laser. Data acquisition was performed with LYSYS II
software, and at least 30,000 events were collected.
Cell sorting of MNCs and double-labeled CD34+ cells was also performed with the FACS Vantage flow cytometer. The labeled cells were suspended at a concentration of 106/mL in PBE buffer, and any residual erythrocytes and dead cells were gated out using forward- and side-scatter channels. A gate was set around the lymphocyte population, and CD34 expression in this gate was assessed. Positively labeled CD34+ cells were then gated and sorted according to CCR1 expression.
The fractions obtained after fluorescein-activated cell sorting were cultured in appropriate assays or stained for morphological evaluation as described in Results.
Colony assays
Sorted CD34+CCR1+ or
CD34+CCR1- cells (1,0002,000) were plated in
a mixture consisting of 1.35% (v/v) methylcellulose (Sigma,
Dorsetshire, England), 30% (v/v) fetal calf serum (Autogen Bioclear,
Calne, England), 1% (v/v) bovine serum albumin (Sigma), 10% (v/v)
5637 conditioned medium from the EJ bladder carcinoma cell
line, and 2 U/mL of EPO to detect GM-CFC and erythroid-forming unit,
bursts (BFU-E) progenitors. The cultures were plated in
triplicate and incubated for 14 days at 37°C in a humidified
atmosphere containing 5% CO2 and 5% O2.
GM-CFCs and BFU-Es were counted according to standard criteria
[29
]. In some experiments, colonies were removed after
14 days, cytospins were prepared, and the cells were evaluated for
morphology after being stained with May-Grünwald/Giemsa.
Serum-free cultures
Cells (n=5,000) from the selected cell populations
described in Results were cultured in X-VIVO 10 serum-free medium
(Bio-Whittaker, Workingham, England) containing 2 mM
L-glutamine and cytokines, which promoted either growth of
erythroid cells or granulocytes and macrophages. Growth of erythroid
cells was stimulated with SCF (100 ng/mL), EPO (5 U/mL), and TPO (50
ng/mL), and cultures to promote granulocytes and macrophages were
supplemented with SCF (100 ng/mL), GM-CSF (10 ng/mL), and IL-3 (10
ng/mL). The cells were incubated for 714 days at 37°C in a
humidified atmosphere containing 5% CO2 and 5%
O2. At days 7 and 14, the cultures were assessed for
progenitor cell content.
Statistical analysis
The Wilcoxon rank sum test and the Wilcoxon signed rank test
were used for paired samples.
|
|
|---|
|
View this table: [in a new window] |
Table 1. Morphological Analysis of CCR1+ and CCR1-
Mononuclear Cells from Cord Blood
|
66% in
the CD34+ bone marrow cells (data not shown).
![]() View larger version (30K): [in a new window] |
Figure 1. Fluorescein-activated cell sorting of CD34+ cells
expressing CCR1. CD34+ cells were sequentially stained with
anti-CCR1 and anti-rabbit-FITC. Cord blood CD34+ cells were
double stained with CD34-PE antibody, and all cells in the lymphocyte
gate were analyzed for CD34 expression. After gating on
CD34+ cells, CCR1 staining was determined. (A) Lymphocyte
gate set on light scatter of CD34 selected cells, (B) CD34-PE
expression determined on cells double labeled with secondary
FITC-conjugated antibody to set the negative CCR1 population, (C) CCR1
negative and positive expression on CD34-PE labeled cells, and (D)
CD34+ cells labeled with FITC and PE control antibodies.
|
|
View this table: [in a new window] |
Table 2. Colonies Generated from CD34+ Cord Blood Cell
Subpopulations Stained with Anti-CCR1 Antibody
|
![]() View larger version (47K): [in a new window] |
Figure 2. Colonies and morphology of cells generated from
CD34+CCR1+ and
CD34+CCR1- cells. Colonies observed when
sorted cell populations were plated in clonogenic assays. (A) Whole
plate and (D) colonies in A at higher magnification; colonies generated
by CD34+CCR1- cells (B) whole plate and (F)
erythroid colonies in B at higher magnification. Colonies from each
plate were pelleted, and methylcellulose was removed (C and E).
Cytospins were prepared from the cell pellets and stained with
May-Grünwald/Giemsa. Macrophages and granulocytes (G) from
CD34+CCR1+ cultures and erythroid cells (H)
from the CD34+CCR1- cultures._art>
|
on colony formation by
CD34+CCR1+ and
CD34+CCR1- cells
enhanced GM-CFC
colony formation on CD34+ cells from cord blood in a
dose-dependent manner. Next, we examined the effect of MIP-1
on
the selected CD34+CCR1+ and
CD34+CCR1-cells. As shown in Figure 3
A, MIP-1
significantly enhanced colony formation by
CD34+CCR1+ cells, with colony numbers
increasing to between 135 and 191% of untreated control values. The
colonies that were formed in the assay by these cells with or without
MIP-1
were composed mainly of granulocytes and macrophages. In the
CD34+CCR1- cultures, the erythroid colony
numbers were reduced to between 68.5 and 98.3% of the untreated
controls in the presence of MIP-1
, as illustrated in Figure 3B
,
showing that the inhibitory response on the erythroid colony-forming
cells was maintained in the absence of the CCR1 receptor.
![]() View larger version (36K): [in a new window] |
Figure 3. Effect of human MIP-1 on colony formation by GM-CFCs in standard
clonogenic assays from CCR1+ and CCR1- cells.
Results shown are from individual experiments with CCR1+
cells (n=6) and CCR1- cells (n=5)
and are expressed as a percentage of the untreated controls. Control
colony numbers ranged from 2653 for GM-CFCs and 87220 for BFU-Es.
There was significant enhancement in colony formation of GM-CFCs in the
presence of MIP-1 (P<0.03).
|
![]() View larger version (24K): [in a new window] |
Figure 4. Phenotypic analysis of cultured CCR1+ and
CCR1- cells. Cells were sorted as indicated in Figure 1C
.
The sorted CCR1+ cells were cultured in serum-free medium
in the presence of SCF, GM-CSF, and IL-3, and CCR1- cells
were cultured in serum-free medium in the presence of SCF, EPO, and TPO
(see Materials and Methods). After 14 days of culture, the cells were
stained with PE-labeled CD71 or PE-labeled glycophorin A and analyzed
by flow cytometry. CCR1+ cultured cells are shown in A and
C, and CCR1- cells are shown in B and D. Control
antibodies (red) and specific staining (black) are
illustrated._art>
|
|
|
|---|
The clonal assay we used allowed the simultaneous development of GM-CFC-originating cells from the granulocyte and macrophage lineages and of BFU-Es, which generated colonies of erythroid cells. Both CD34+CCR1+ and CD34+CCR1- cell fractions were examined in this assay, and the findings that CD34+CCR1- cells predominantly gave rise to erythroid colonies by 14 days in culture suggested that the BFU-Es are mainly contained in the CD34+CCR1- cells. In line with this finding, CD34+CCR1- cells produced erythroblasts in liquid suspension cultures, and these cells expressed high levels of CD71, the human transferrin receptor, and glycophorin A, another marker of the erythroid lineage. Moreover, >96% of the GM-CFCs segregated to the CD34+CCR1+ cell fraction and generated granulocyte/macrophage colonies (Table 2) . Proliferation of these cells in liquid culture was also extensive, and macrophages and immature neutrophils were apparent at 14 days. The enrichment of colonies in the two subfractions indicated that these isolated populations are biologically distinct in lineage commitment programs. The distinction between the erythroid and monomyelocytic lineages was also seen in freshly isolated cord blood cells, when expression of CCR1 was examined on MNCs (Table 1) . Thus, CCR1 is mainly expressed on immature and mature cells of the granulocyte and macrophage lineages, and cells destined to develop (BFU-Es) and those already developing along the erythroid lineage exhibit at best low levels of the receptor.
Some GM-CFCs were also found in the CD34+CCR1- population, which might reflect insufficient separation of this subset and the necessity for a more stringent gating strategy. In the gating strategy depicted in Figure 1 , we designated the cells as CD34+CCR1- rather than CD34+CCR1low because, although there might be a few negative cells in the positive cell population, there are no positive cells in the negative cell population. Indeed, in three separate experiments, when the cell populations exhibiting the lowest or the highest levels of immunofluorescence were isolated, there was better discrimination between erythroid and granulocyte/macrophage progenitors. In fact, 99% of erythroid colonies were obtained in the CD34+CCR1- fraction, with 96% granulocyte/macrophage colonies in the CD34+CCR1+ subset. Therefore, we suggest that CCR1- designates a population of cells that do not express the receptor.
Primary human erythroid progenitors are difficult to obtain as
homogeneous cell populations, and various strategies for isolation or
enrichment have been reported. These include differential density
centrifugation and negative and positive selection by immunomagnetic
separation with purities ranging from
1856% [1
2
3
,
5
]. Alternative methods include ex vivo culture of MNCs
or CD34+ cells in defined growth factor combinations to
generate large numbers of specific progenitors, although contamination
by mature cells that also develop in the cultures is often a problem
[5
, 30
]. Two recent studies distinguished
GM-CFCs and BFU-Es on the basis of their ability to bind biotinylated
c-kit ligand or multiple cell surface antigens
[31
, 32
]. The results of these studies
compare favorably with results in the present study, in which 97% of
the GM-CFCs were present among CD34+CCR1+
cells, with 81% of BFU-Es in the CD34+CCR1-
fraction. Monoclonal antibodies against a number of other cell surface
antigens including c-kit, CD71, HTK, and the EPO receptor
have been generated [4
, 33
34
35
]. All can
distinguish erythroid precursors and have similar patterns of
expression, in that their expression reaches maximum at the BFU-E stage
and declines during terminal differentiation. CCR1 is distinct in that
expression is absent from the progenitor stage onwards.
CD34+ cells expressing high levels of the receptor are
restricted to the granulocyte/macrophage lineages.
The exact expression pattern of CCR1 during hemopoietic cell
differentiation is not known, and some conflicting results have been
reported. Although mRNA for CCR1 has been detected in CD34+
cells in cord blood and also in bone marrow using reverse
transcriptase-PCR and Northern blotting [14
,
26
, 36
], Majka was unable to demonstrate the
presence of CCR1 on CD34+ cells in human bone marrow MNCs
[37
]. Previously published data indicated that CCR1 is
present on mature erythroblasts and erythroid progenitors in bone
marrow [38
], but the data presented here show that this
is not the case with cord blood. It is possible that expression of CCR1
varies depending on the sample source, because we noted a differential
response to MIP-1
depending on the source of the hemopoietic cells
[14
].
MIP-1
enhanced colony formation of granulocyte-macrophage
progenitors by the CD34+CCR1+ cells (Figure 3)
,
suggesting that the proliferation response is mediated by the CCR1
receptor. Erythroid progenitors, which are CCR1-,
maintained an inhibitory response, as shown by a decrease in formation
of erythroid colonies, indicating that this inhibitory effect is
mediated by a MIP-1
receptor other than CCR1. This agrees with
compelling evidence that the CCR1 receptor might mediate the
proliferative response to MIP-1
. In bone marrow cells from mice with
a targeted disruption of CCR1, MIP-1
failed to enhance colony
formation in clonogenic assays stimulated with GM-CSF or macrophage
colony-stimulating factor. In contrast, marrow cells from wild-type
animals responded normally with enhanced proliferation in the presence
of MIP-1
[39
]. The antibody used did not seriously
interfere with the function of the cells in the biological assays,
given the fact that the positively labeled cells (CCR1+)
were able to respond effectively when MIP-1
was added.
Simplified methods for isolation of specific lineage-committed progenitors would be advantageous for a variety of applications. It would allow for definitive studies of selective expression of specific genes at distinct stages of differentiation and commitment as well as examination of genes whose targeted disruptions lead to defects in myeloid lineages. The procedure outlined here using the anti-CCR1 antibody should prove useful in distinguishing the myeloid lineages.
Received January 2, 2001; accepted January 17, 2001.
|
|
|---|
and transforming growth factor ß on hematopoietic progenitor/stem cell growth Blood 84,2175-2181This article has been cited by other articles:
![]() |
M. He, R. Horuk, S. M. Moochhala, and M. Bhatia Treatment with BX471, a CC chemokine receptor 1 antagonist, attenuates systemic inflammatory response during sepsis Am J Physiol Gastrointest Liver Physiol, April 1, 2007; 292(4): G1173 - G1180. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Lu, Y. Nakamoto, Y. Nemoto-Sasaki, C. Fujii, H. Wang, M. Hashii, Y. Ohmoto, S. Kaneko, K. Kobayashi, and N. Mukaida Potential Interaction between CCR1 and Its Ligand, CCL3, Induced by Endogenously Produced Interleukin-1 in Human Hepatomas Am. J. Pathol., April 1, 2003; 162(4): 1249 - 1258. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. G. Manz, T. Miyamoto, K. Akashi, and I. L. Weissman Prospective isolation of human clonogenic common myeloid progenitors PNAS, September 3, 2002; 99(18): 11872 - 11877. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||