Note: Descriptions are shown in the official language in which they were submitted.
CA 02219869 1997-10-31
Human CD34- Hematopoietic Stem Cells
Field of the Invention
The present invention relates to human hematopoietic stem cells and to a
method of isolating and using such cells.
Background of the Invention
The m~mm~ n hematopoietic system consists of a heterogeneous array
of cells ranging from large numbers of dirrelellliated cells with defined function
to rare pluripotent stem cells with extensive developmental and proliferative
potential (1, 2, 3). The defining feature of a stem cell is its ability to repopulate
the hematopoietic system of a recipient after transplantation. Stem cells are
playing an increasingly important role in clinical and commercial applications,
as the role of the stem cells in transplantation widens. Identification and
purification of stem cells is essential both to determine the cellular and
molecular factors that govern stem cell development and for the application of
clinical procedures including stem cell transplant and gene therapy.
Cell surface expression of CD34 has become the distinguishing feature
used to isolate stem cells because CD34 is downregulated as cells differentiate
into more abundant mature cells (4). However, CD34 does not mark stem cells
exclusively since only 1% of bone marrow (BM) cells are CD34+ and include
clonogenic progenitors that are not able to repopulate after transplantation.
Therefore other markers such as Thy-l can be combined with CD34 to
positively select for a more enriched cell fraction (5, 6, 7). Conversely, the
CD34+ fraction can be enriched by elimin~ting cells that express markers that
are expressed on non-repopulating cells (e.g.lineage antigens). For these
reasons, all current clinical and experimental protocols ~tili~ing human stem
cells including ex vivo culture, gene therapy, and bone marrow transplantation,
focus on CD34+ cells.
CA 02219869 1997-10-31
There have been reports (8, 9, 10) of murine CD34- hematopoietic stem
cells which are capable of long term repopulation. For human hematopoietic
stem cells, however, CD34+ cells have been regarded as a stem cell marker
without exception.
Brief Description of the Drawings
The present invention will be further understood from the following
description with reference to the Figures, in which:
Figure lA demonstrates cell surface expression of CD34 on cord blood
cells depleted for lineage markers (Lin-). (Panel I) Lin- cells were purified as in
(11, 12) and stained with a class III monoclonal antibody for CD34 (581)
conjugated to FITC (Becken Dickenson, BD). Cells residing in R1 were
considered CD34 negative (CD34-). (Panel II) CD34- cells were purified using
standard cell sorting techniques and re-analyzed using the same CD34-581
antibody. (Panel III and Panel IV) Purified R1 cells were stained and re-
analyzed using a class I monoclonal antibody for CD34 (Immun-133) (Coulter)
and a class II monoclonal antibody for CD34 (Q-Bend-10) (Becton Dickinson,
BD).
Figure lB demonstrates the immunostaining of purified populations for
CD34 expression. Representative cells are shown from a total of 25-75 cells
examined for each treatment (n=2).
Figure lC demonstrates a comparison of cell surface markers between
Lin-CD34- versus Lin-CD34+ cord blood cells commonly used to further
subdivide p~ ive cells.
Figure 2 demonstrates the level of human cell engraftment in
NOD/SCID mice transplanted with highly purified Lin-CD34- cells at various
doses.
Figure 3 demonstrates the multilineage differentiation of human Lin-
CD34- cells in NOD/SCID mice. Bone marrow from a highly engrafted mouse
transplanted with 120,000 Lin-CD34- cord blood cells was stained with various
CA 02219869 1997-10-31
human-specific monoclonal antibodies and analyzed by flow cytometry.
Approximately 106 mononuclear cells collected from mouse BM was prepared
as shown previously for multilineage analysis (12). (A) Cells with medium to
high forward scatter (region Rl) were gated and further analyzed. (B)
Histogram of CD45 (pan-leukocyte marker) expression indicating that 2.5% of
the cells present in the murine bone marrow are h~ n, gated R2. All further
lineage markers was done on cells within gate R2 (CD45+). (C) Isotype control
for non-specific IgG staining of PE and FITC fluorescence. (D) Expression of
myeloid marker CD33 and granulocyte marker CD15; (E) pan-B cell markers
CDl9 and CD20; (F) CD38 and the imm~h1re hematopoietic marker CD34; (G
and H) T-cell markers CD2, CD3, CD4 and CD8.
Figure 4 is a frequency analysis of Lin-CD34- cells found in human
hematopoietic tissue. (A) Column I; Fetal liver collected from 8 week old
human fetus, n=3 Column II; Fetal blood aspirated from 19 week old fetus, n=l
Column III; Cord blood collected from placenta at time of birth, n=3. (B)
Column I; Normal Adult Bone marrow, n=2 Column II; Bone marrow from a
normal adult donor after 5 days of G-CSF ~lmini~tration, n=2 Column III;
Peripherial blood collected from a normal adult donor after 5 days of G-CSF
a-lministration, n=2.
Figure 5 demonstrates the percentage of the input cells after in-vitro
culture of lin- CD34- cells. Purified cells were counted and seeded (250-2000)
in wells cont~ining serum free media (SF) ( see examples) (solid bar) or SF
supplemented with 25% HWEC- conditioned media ( shaded bar). Cells were
counted each day and the percentage of the input cells was calculated ( n=3).
Figure 6 is an analysis of CD34 and CD38 expression of highly purified
populations after in vitro culture. A representative experiment (n=4) of CD34
and CD38 cell surface expression performed on initially purified lin-CD34-
CD38- cells, and purified cells after 2 and 4 days of culture in SF, SF
supplemented with 5% FCS or 25% HWEC-CM. The entire contents of
individual wells was collected at 2 and 4 days (5000-10, 000 cells), stained with
CA 02219869 1997-10-31
monoclonal antibodies CD34 and CD38 directly conjugated to FITC and PE
respectively (Beckton Dickinson, BD). Stained populations were then washed
and analyzed using standard flow cytometric techniques as done previously
(11,12) followed by the display of histograms using the Cell Quest software
program (BD)
Figure 7 demonstrates the capacity of expanded lin-CD34-, lin-CD38-
CD38- or lin-CD34-CD38+ cell fractions to engraft NOD/SCID mice.
Figure 8 demonstrates the multilineage differentiation of human lin-
CD34- cells in NOD/SCID mice after ex vivo culture. (A) Histogram of CD45
(human-specific pan-leukocyte marker) expression indicating that 7% of the
cells present in the murine BM are hllm~n (B) Forward and Side scatter of the
CD45 human cells. Subsequent analysis of lineage markers was done on
CD45+ cells within gate Rl (lymphoid and blast cells) or R2 (myeloid cells)
gates. (C) Analysis for the presence of imm~tllre cells using the CD34 and
CD38 markers. ( D) Analysis for the presence of human B cell lineage cells
using CDl9 and CD20 markers. (E-F) Analysis for the presence of human T
lymphocytes using the panel of T cell markers: CD2, CD3, CD4 and CD8. (G-
H) Analysis for the presence of myeloid cells using CD33, CD14, CD15 and
CD13 markers.
Description of the Invention
The present inventors have identified and isolated a population of human
hematopoietic stem cells which do not express CD34 (CD34-), CD38 (CD38-)
or lineage specific markers (Lin-) and which are able to generate by
proliferation and dirrelellLiation multiple lineages of the human hematopoietic
system, as evidenced by their ability to produce multilineage human
engraftment of immune-deficient NOD/SCID mice after transplantation.
Moreover, the repopulative capacity and the dirrerellliative capacity of the lin-
CD34-CD38- cells can be stimulated by in vitro culture of these cells.
CA 02219869 1997-10-31
In initial studies, a population of CD34-Lin- cells were isolated from
human fetal liver, cord blood, bone marrow and mobilized peripheral blood and
bone marrow. These cells do not express any lineage specific markers and are
also devoid of HLA-DR and Thy-l, two other markers associated with
primitive cells, but are heterogeneous for CD38 and c-kit expression. The
NOD/SCID transplantation studies show that the repopulating cells are
exclusively present in the CD38- fraction and not the CD38+ fraction and these
cells are characterized as Lin-CD34-CD38-. These novel Lin-CD34- cells are
most abundant in fetal tissues, with progressive reductions in cord blood (CB)
and adult bone marrow (BM) or mobilized peripheral blood (M-PB).
Transplantation of the Lin- CD34- cells in NOD/SCID mice engrafted myeloid,
lymphoid and erythroid lineages, which demonstrates that Lin- CD34- cells
repopulate cells of the hematopoietic system. The results also show that Lin-
CD34-CD38- cells are developmentally earlier than CD34+ cells and can
produce CD34+ cells following in vitro culture or after repopulation. Thus,
these cells appear to be at the top of the heirarchical structure of human stem
cells.
The identification of this novel stem cell within the hierarchy of human
hematopoiesis has important implications for understanding the origin of
hematopoietic diseases such as leukemia and for clinical procedures such as
stem cell transplantation and gene therapy. This novel stem cell can be
important for the treatment of infection, for the reconstitution of deficient ormissing cell populations as for example for cancer patients after myeloablative
therapy and for the treatment of genetic abnormalities and defects.
The present invention provides a therapeutic composition and a method
for the treatment of hematopoietic disorders. In particular, the composition andmethod can be used to treat hematopoietic disorders such as leukemia and for
several clinical procedures such as stem cell transplantation, therapy, gene
therapy, for combating infection and for cell reconstitution.
CA 02219869 1997-10-31
In accordance with an aspect of the present invention there is provided
an essentially pure population of human hematopoietic cells characterized as
Lin- CD34-.
In accordance with another aspect of the present invention there is
provided an essentially pure population of human hematopoietic cells
characterized as Lin-CD34-CD38-.
In accordance with another aspect of the present invention there is
provided a method for providing an essentially pure population of human
hematopoietic cells characterized as Lin- CD34-.
In accordance with another aspect of the invention there is provided a
method for the treatment of hematopoietic disorders such as leukemia
comprising the use of human hematopoietic cells characterized as Lin- CD34-.
In accordance with another aspect of the present invention is a method
for the ex vivo generation of human hematopoietic cells using Lin-CD34- cells.
In accordance with another aspect of the present invention is a gene
therapy method for providing genetically altered human hematopoietic cells
characterized as Lin- CD34- to a patient.
In accordance with another aspect of the present invention is a method
for reconstituting deficient or missing human hematopoietic cell populations
comprising the use of Lin- CD34- cells.
In accordance with another aspect of the present invention is a method
for transplanting human hematopoietic cell populations to a patient comprising
the use of Lin- CD34- cells.
In accordance with another aspect of the present invention is a method
for combatting infection in a patient comprising ~dmini~tering an effective
amount of human hematopoietic cells characterized as Lin- CD34-.
In accordance with yet a further aspect of the present invention there is
provided a method utilizing human hematopoietic cells characterized as Lin-
CD34- for the production of CD34+ human hematopoietic cells.
CA 02219869 1997-10-31
In accordance with a further aspect of the present invention is a method
for increasing the repopulating capacity of human hematopoietic cells
characterized as Lin-CD34- by culturing such cells in vitro for several days.
In accordance with a further aspect of the present invention is a method
for screening candidate compounds affecting proliferation or dirr~relltiation ofstem cells characterized as Lin-CD34- and Lin-CD34-CD38-.
It was determined using CD34 class III fluorescent monoclonal
antibodies that CD34- cells that do not express lineage markers exist in human
hematopoietic tissues (Fig. lA). The possibility that CD34 protein was
produced in the Lin-CD34- cells but not transported to the cell surface, was
excluded using permeabilized, stained cytospins of purified Lin-CD34- that
were further conjugated with fluourescent monoclonal antibodies. This
confirmed that no CD34-FITC signal was detected in the Lin-CD34- cells (Fig.
lB)
Heterogeneity within human Lin-CD34+ cells is well documented and
further subdivision for the most p~ ive cells is typically based on the cell
surface markers CD38, c-kit, Thy-1 and HLA-DR. The expression of these
markers on both the Lin-CD34+ and Lin-CD34- cells was compared (Fig. lC).
Lin-CD34- cells displayed a bi-modal distribution of CD38 clearly dividing the
population into two fractions in contrast to the high proportion of Lin-CD34+
cells that express CD38 (Fig. lC). Cell surface expression of c-kit was similar
between that two populations, while the Lin-CD34- cells are almost exclusively
Thy-1- and HLA-DR- (Fig. lC). Both the absence of HLA-DR expression and
the presence of Thy-1 have been proposed to define more ~lhlliLive
subfractions within the Lin-CD34+ population. Therefore, the Lin-CD34-
population derived from CB is a distinct population which differs not only in
CD34 expression from plilllilive Lin-CD34+ cells but also in phenotypic
heterogeneity based on additional markers associated with stem cells.
Using clonogenic methylcellulose assays, it was determined whether
Lin-CD34-, Lin-CD34-CD38- and Lin-CD34-CD38+ cells possessed any
CA 02219869 1997-10-31
hematopoietic progenitor activity by comparing the CFC and LTC-IC content,
respectively. Clonogenic capacity of Lin-CD34- cells was extremely low in
comparison to Lin-CD34+CD38- cells (Table I). As many as 10,000 cells
needed to be seeded on MS-5 stroma to detect a single LTC-IC within the Lin-
CD34- cell fraction, while further purification demonstrated that detection of
LTC-IC in the Lin-CD34-CD38- fraction required seeding of at least 2000
cells. By contrast, as few as 10 Lin-CD34+CD38- cells contain an LTC-IC. The
Lin-CD34-CD38+ cells were devoid of LTC-IC activity (limit of detection at
10,000 cells) but contained a much higher capacity to form CFC specifically
commilled to the erythroid lineage (Table I). The low efficiency to produce
myeloid and erythroid colnlllilled progenitors as well as the more plilllilive
LTC-IC is similar to observations made with murine Lin-CD34- cells in the
same assay systems and suggest functional similarities may exist between the
Lin-CD34- cells from these two species (8,10).
The only conclusive method to detect stem cells is to determine their
ability to repopulate recipient hosts (1, 2,3). A repopulation assay was
developed for primitive human cells based on their ability to initiate
multilineage human engraftment in immune-deficient NOD/SCID mice (13,
14). Based on cell purification and gene marking, the cell capable of
repopulating NOD/SCID mice (termed the SCID-Repopulating Cell, SRC) was
established as distinct from and more plilnilive than the majority of progenitors
detected in in vitro assays (15). Only the Lin-CD34+CD38- cells and not the
Lin-CD34+CD38+ cells could give rise to multilineage engraftment (15, 12).
Moreover, transplantation of as many as 10-6 Lin+CD34- cells also did not
engraft (12). To determine whether highly purified Lin-CD34- cells had SRC
activity and to determine the frequency of any repopulating cells, Lin-CD34-
cells were transplanted at varying cell doses into NOD/SCID mice using our
standard protocols, and BM was analyzed for the presence of human cells after
8-12 weeks. The level of human cell engraftment in 23 NOD/SCID mice was
quantitated by FACS and DNA analysis for the presence of human cells and
CA 02219869 1997-10-31
results are sllmm~ri~ed in Fig. 2. A large proportion of transplanted mice were
engrafted with human cells indicating that the Lin-CD34- cells were able to
repopulate NOD/SCID mice. We have termed this cell the CD34NEG-SCID
Repopulating Cell (CD34NEG-SRC). The frequency was 1 CD34NEG-SRC in
125,000 Lin-CD34- cells. The differentiative and proliferative capacity of the
CD34NEG-SRC cell was assessed by flow cytometric analysis. A
representative engrafted NOD/SCID mouse 10 weeks after the transplant of
120,000 Lin-CD34- cells is shown in Fig.3. Cells with medium to high forward
scatter (region R1, Fig. 3A) were gated and further analyzed based on CD45
expression, a human specific pan-leukocyte marker (Fig.3B). The isotype
control is shown in Fig.3C. The BM of this mouse contained 2.5% CD45+
human cells (Fig.3B) or at least 106 total human cells indicating that the Lin-
CD34- cells have extensive proliferative capacity. Granulocytes (CD15+) were
present among myeloid cells (CD33+) (Fig.3D). B-lymphoid cells were also
present in the murine BM as shown by staining for CD19 and CD20 (Fig. 3E).
Interestingly, human T-cells expressing both CD2 and CD3 (Fig. 3G), along
with CD4 and CD8 positive cells (Fig.3H) were also identified. NOD/SCID
mice transplanted with highly purified p~ ilive Lin-CD34+CD38- cells never
gave rise to engraftment contailling T-cells demonstrating the unique in vivo
repopulation behavior of the Lin-CD34- cells. In addition to multilineage
engraftment, imm~ re CD34+ and CD34+CD38- cells were detected (Fig.3F).
It was concluded that the Lin-CD34- cells have the ability to repopulate
NOD/SCID mice and differentiate in vivo into multiple lineages of myeloid and
lymphoid cells. The production of CD34+CD38- and CD34+CD38+ cells in
vivo suggests that Lin-CD34- cells are developmentally earlier than CD34+
cells in the hierarchy of human hematopoiesis.
There is evidence that the frequency of plilllilive cells changes during
ontogeny with the highest proportions seen in the fetus. A variety of fetal,
neonatal, and adult sources of human hematopoietic tissue were analyzed in an
attempt to identify and quantify the Lin-CD34- population. The results indicate
CA 02219869 1997-10-31
(Fig.4A,4B) that Lin-CD34- cells are produced early in human ontogeny and
can persist throughout adult life and that the mechanisms that operate during
the mobilization of CD34+ cells with G-CSF also affect Lin-CD34- cells.
This data provides the first identification of a novel human
hematopoietic stem cell that does not express CD34 or lineage-specific
markers. As d~te~ ed by all available monoclonal antibodies, this population
is not only distinct by the absence of CD34, but also by the lack of HLA-DR
and Thy- 1 markers. In addition to these phentotypic differences, several lines of
evidence functionally distinguish these two stem cell populations. While the
Lin-CD34- cells have limited hematopoietic activity in vitro, Lin-CD34+CD38-
cells are highly clonogenic based on their ability to produce CFC and LTC-IC.
Although both stem cell fractions are capable of repopulation, the presence of
T-cells within the multilineage engraftment is a unique characteristic of Lin-
CD34- transplantation since we have never detected human T cells in mice
transplanted with Lin-CD34+CD38- cells (n=25). Therefore it is unlikely that
NOD/SCID repopulation is derived from the co~ "~ tion of Lin-
CD34+CD38-cells. Furthermore, based on LTC-IC frequency a millill~Ulll of 1
LTC-IC resides within 10 highly purified Lin-CD34+CD38- cells. In contrast to
flow cytometry, in which we have determined that the Lin-CD34- population is
99% (or in some cases 100%) pure, the LTC-IC assay allows us to detect a
smaller number of col~ ting Lin-CD34+CD38- cells. Using this assay,
only a single LTC-IC could be detected in as many as 10 000 CD34-Lin- cells.
If this LTC-IC activity came from a Lin-CD34+CD38- cell, a maximum of 10
Lin-CD34+CD38- cells could be contained in the Lin-CD34- purified fraction
(0.1% cont~min~tion). In addition, repopulated mice have not been observed
when only 10 Lin-CD34+CD38- cells were transplanted (12). Since the
frequency of SRC derived from Lin-CD34- cells is 1 in 125 000, a maximum of
125 Lin-CD34+CD38- cells could have been transplanted, again less than the
number needed to repopulate CD34+CD38- cells.
CA 02219869 1997-10-31
The in vivo differentiation of human Lin-CD34- cells into CD34+ and
lineage positive cells suggests that the Lin-CD34- cells preceed CD34+ cells in
the hierarchy of human hematopoiesis.
The identification of these novel repopulating cells, termed CD34neg-
SCID repopulating cells (CD34neg-SRC), provides an OppOl lUllily to examine
the differentiation and proliferation potential of these cells and to establish their
relationship to other cells within the human stem cell hierarchy. However, in
vivo repopulation is a complex system m~king it difficult to establish
hierarchical relationships. Moreover, it was found that there is 1 CD34neg-SRC
in 105 Lin-CD34- cells. While this may reflect the true frequency of primitive
cells capable of repopulation, it it also possible that other plilllilive cells exist
within this cell fraction. The inability to demonstrate activity of these plilllliive
cells could be due to limitations in currently available hematopoietic assays.
For example, these cells could possess the developmental potential to
repopulate NOD/SCID mice but are unable to under the current transplant
conditions (e.g. some repopulating cells have much slower kinetics, or may lack
expression of a particular adhesion molecule, etc) and therefore are never
detected. In the murine system, some stem cells will only repopulate long-term
but not short-term and vice versa (1,2,3). Therefore, ex vivo cultures are better
suited for studies that examine the proliferative and differentiative potential of
primitive cells, particularly in response to cytokine stimulation, because the
conditions are well defined and it is straightforward to (measure developmental
changes). Many studies based on ex vivo culture have demonstrated that there
is heterogeneity within the CD34+ cell fraction (4). Subfractionation of the
CD34+ cells on the basis of Thyl, CD38, and HLA-DR expression together
with in vitro clonogenic and LTC-IC assays have demonstrated the progenitor-
progeny relationship of the various cell types that make up the stem cell
hierarchy (4, 14). The availability of the SRC assay to detect even earlier celltypes has added more information about the org~ni7~tion of cells within this
hierarchy (14). Moreover, it was demonstrated that the SRC (derived from
CA 02219869 1997-10-31
CD34+CD38- cells) can be expanded for 4 days in serum-free cultures without
inducing their differentiation. However, all SRC are lost within an additional 4days of culture conco-ll-llilallt to the appearance of more differentiated CD38+cells (11). At the same time, both colony-forming cells (CFC) and long-term
culture initiating cells (LTC-IC) could be greatly expanded during 8 days of
culture demonstrating that the majority of the SRC are a distinct population
(11), but may be closely related to ELTC-IC (18). Thus, in vitro culture systemscan be used to identify very fine transitions in the developmental program by
combining both flow cytometry and functional CFC, LTC-IC and SRC assays.
It is now demonstrated hel~wilh that ex vivo culture of Lin-CD34- cells can
induce the appearance of CD34+ cells and can increase the proportion of Lin-
CD34- cells that have SRC activity. These studies provide new insight into the
developmental program of human hematopoietic stem cells.
Our earlier studies demonstrated that the Lin-CD34- cell fraction
expressed a bimodal distribution of CD38 allowing for further purification into
CD38- and CD38+ subpopulations (12). To determine whether the Lin-CD34-,
Lin-CD34-CD38-, Lin-CD34-CD38+ cells could be induced to proliferate
andlor differentiate, these cells were cultured in defined serum-free (SF)
conditions that have been previously shown could expand Lin34+38- cells and
maintain and modestly increase POS CD34 POS-SRC(ll, 17). Cells were
plated in methlycellulose assays at day 0, and after day 4 of liquid culture
(Table II) to determine the effect of cytokine stimulation on the clonogenic
progenitors present in lin-CD34-, lin-CD34-CD38- and lin-CD34-CD38+ cell
fractions. Both the lin-CD34- and more purified lin-CD34-CD38- cells have a
low plating efficiency (PE), l in 89 and 1 in 297 CFC respectively, whereas a
higher PE of lin-CD34-CD38+ cells 1 in 10.4 cells was seen. Interestingly, the
clonogenic capacity of the lin-CD34-CD38+ cells is restricted to the erythroid
lineage. After 4 days of culture in SF media or SF media supplemented with the
addition of 25% conditioned medium obtained from primary human umbilical
vein endothelial cells (HWEC-CM), the PE of all the sub-populations
CA 02219869 1997-10-31
examined had decreased, whereas the addition of 5% of FCS increased the
clonogenicity of lin-CD34- cells and to a greater extent lin-CD34-CD38+
(Table II). This difference in clonogenicity may reflect heterogenity within Lin-
CD34- cells and demonstrates that the CD38+ subfraction is already co~ llilled
to the erythroid lineage suggesting that the CD34neg-SRC resides in the Lin-
CD34-CD38- subfraction.
To determine whether the lin-CD34-CD38- cells could be stimulated to
proliferate, changes in cell number were recorded between day 0 and 4 of
culture with SF or 25% HWEC-CM (Fig.5). The total number of cells
decreased by 2 fold at day 4 in SF media. Supplementation of 5% fetal calf
serum (FCS) showed no increase in the viability of these cells (data not shown).However, the addition of HWEC-CM to SF media maintained or slightly
increased the total cellularity (Fig.5). These results demonstrate that culture
conditions that are optimal for Lin-CD34+CD38- cells (11), are unable to
support lin-CD34-CD38- cells, while soluble components present in primary
HWEC-CM seem to permit their survival.
The effect of culture on the differentiatiation program of lin-CD34-
CD38- cells, from individual wells was analyzed by flow cytometry after 2 and
4 days of culture in various conditions (Fig.6). Surprisingly, Lin-CD34-CD38-
cells seeded in SF media began to express CD34 which could be enhanced with
the addition of 5% serum (Fig.6). In contrast, the majority of cells obtained
after 2 or 4 days of culture in the presence of 25% HWEC conditioned
medium still maintained the original Lin-CD34-CD38- phenotype. The
stimulation of Lin-CD34-CD38- cells to differentiate and produce
CD34+CD38- cells suggests that precede the CD34+CD38- population in the
hierarchy of human hematopoiesis. Moreover, these results indicate that the lin-CD34-CD38- cells respond to signals present in SF or 5% serum conditions and
that HWEC-CM can inhibit this stimulation.
To confirm which fraction contained CD34neg-SRC, both Lin-CD34-
CD38- and Lin-CD34-CD38+ cells were transplanted into NOD/SCID mice.
CA 02219869 1997-10-31
Transplantation with as few as 10,000 or 4,000 Lin-CD34-CD38- cells derived
from CB or BM, respectively resulted in engraft (Fig. 7), whereas as many as
180,000 Lin-CD34-CD38+ cells were not capable of repopulation (Fig 7).
These data indicate that the CD34neg-SRC present in the Lin-CD34- fraction
are restricted to the CD38- subfraction. However, since an entire CB sample
contains only 1 or 2 CD34neg-SRC (e.g. frequency is 1 CD34neg-SRC in
125,000 Lin-CD34- cells and one CB sample contains up to 250,000 Lin-
CD34- cells) the losses associated with subselecting based on CD38 expression
result in only 9% of samples which repopulate NOD/SCID mice.
As in vitro culture of lin-CD34-CD38- cells caused developmental
changes recognized by the appearance of CD34+ cells, we evaluated whether
the repopulating activity ofthe Lin-CD34-, lin-CD34-CD38-, and lin-CD34-
CD38+ cell fractions was affected by ex vivo culture (Fig 7, panel II). Purifiedfractions from 43 CB and 3 BM samples were cultured for 2 and 4 days in SF,
SF supplemented with 5%FCS or 25% HWEC-CM and transplanted at various
doses into 136 recipient NOD/SCID mice (Fig. 7, Panel II). A total of 13 out 29
mice transplanted with less than 125,000 to as few as 8,700 Lin-CD34- cells
that had been cultured for 4 days were engrafted. A large increase in the
number of CD34neg-SRC must have occurred since 125,000 uncultured Lin-
CD34- cells were required to repopulate NOD/SCID mice. Similarly, 35% of
the mice transplanted with cultured lin-CD34-CD38- cells were engrafted as
compared to 9% of mice transplanted with similar doses of uncultured lin-
CD34-CD38- ( Fig. 7, Panel I vs II). By contrast, 5,000 and 10,000 lin-
CD34+CD38- CB cells, known to contain 10-20 SRC, cultured for 4 days in the
presence of serum or the addition of 25% HWEC-CM were unable to engraft
NOD/SCID mice, confirming the absence of collt~ ting SRC from this
fraction in Lin-CD34-CD38- repopulating cells. HWEC-CM provided
significant increases in the proportion of repopulating cells. Consistent with the
inability of lin-CD34-CD38+ cells to engraft at day 0, mice transplanted with
14
CA 02219869 1997-10-31
cultured cells were not engrafted (limit of detection < 0.05%) confinning the
absence of repopulating cells within this fraction.
The BM of engrafted mice was analyzed by multiparameter flow
cytometry to determine whether cultured Lin-CD34- repopulating cells
possessed the same in vivo proliferative and differentiative capacity as
uncultured cells. A representative analysis of the BM of a NOD/SCID mouse
transplanted with an initial population of 40,000 lin-CD34- cells after 4 days of
culture is shown (Fig 8). The BM of this mouse contained 7% human cells as
detected by CD45 expression, a human specific pan-leukocyte marker (Fig. 8).
Both B and T-lymphoid cells were present in the murine BM as shown by
staining for CDl9, CD20 and CD4, CD3 antigens (Fig. 8D-F). The presence of
CD33+, CD14+, CD15+ and CD13+ cells indicated the differentiation potential
of lin-CD38-CD38- cells to the myeloid lineages (Fig. 8G-H). The engraftment
pattern of mice transplanted with expanded lin-CD34-CD38- cells is similar to
that observed with unstimulated purified lin-CD34- cells. The presence of
human T-cells is a unique feature of Lin-CD34- engraftment since T-cells have
neither been detected in mice transplanted with purified Lin-CD34+CD38- cells
before or after ex vivo culture (11, 17).
It is believed that the repopulating cells of the human hematopoietic
system are the lin-CD34-CD38- sub-population and short term ex vivo culture
of this fraction has been observed to increase the proportion of repopulating
cells. The ability of Lin-CD34-CD38- cells to produce CD34+CD38- cells in
vitro and in vivo, demonstrates the developmental capacity of these cells and
further suggests that this population of cells is more primitive than the CD34
positive fraction. In addition, conditions evaluated here provide the foundationfor future gene transfer and ex-vivo expansion of this novel population and for
the identification of factors that stimulate their proliferation and dirreretlliation.
Furthermore, the knowledge that Lin-CD34- cells exist and can
repopulate human hematopoietic cells provides a novel therapeutic composition
and a method for the treatment of hematopoietic disorders. In particular, the
CA 02219869 1997-10-31
composition and method can be used to treat hematopoietic disorders such as
leukemia and for several clinical procedures such as stem cell transplantation,
therapy, for combating infection and for cell reconstitution. These cells can
also be used to generate CD34+ cells.
Examples
The examples are described for the purposes of illustration and are not
intended to limit the scope of the invention.
Methods of synthetic chemistry, protein and peptide biochemistry and
immunology referred to but not explicitly described in this disclosure and
examples are reported in the scientific literature and are well known to those
skilled in the art.
Example 1 - Analysis of Lin-CD34- cells found in human hematopoietic tissue
Mononuclear cells were isolated from various human hematopoietic cell
sources and stained with monoclonal antibodies for CD2, CD3, CD4, CD7,
CD13, CD14, CD15, CD16, CDl9, CD20, and glycophorin conjugated to
FITC, CD38 conjugated to PE and CD34 conjugated to Cy-5 . Cells gated Rl
did not express lineage associated markers (Lin-) and were further analyzed for
the expression of CD34 and CD38.
Identification of CD34- cells with no lineage markers in human hematopoietic
tissues
To determine whether CD34- cells that do not express lineage markers
exist in human hematopoietic tissues, human cord blood (CB) cells were first
depleted of mononuclear cells that express 15 different lineage-specific
antigens from human cord blood. This Lin- population was 99% pure (data not
shown). The Lin- cells were then stained with the most widely used CD34 class
III monoclonal antibody conjugated to FITC. Flow cytometric analysis showed
two distinct populations of CD34+ and CD34- cells (Fig lA panel 1). The
CA 02219869 1997-10-31
CD34- cells (gated Rl, Fig. lA, I) were collected by flow sorting, reanalysis
demonstrated their high purity (99%; Fig. lA, II). To confirm that these cells
did not express any surface CD34 antigen, the sorted cells were re-stained with
two other CD34 Class I and II antibodies that recognize different epitopes of
the CD34 molecule (Fig. lA, III and IV). No CD34+ cells were detected,
attesting to the high purity and lack of CD34 cell surface expression of this Lin-
CD34- population.
To exclude the possibility that CD34 protein was produced in the Lin-
CD34- cells but not transported to the cell surface, cytospins of purified Lin-
CD34- were permeabilized, stained with CD34 monoclonal antibodies
conjugated to FITC and counter stained with DAPI (Fig. lB). No CD34-FITC
signal was detected in the Lin-CD34- cells. The specificity of the procedure is
shown by the detection of cell surface and intracellular expression of CD34 on
a population of purified Lin-CD34+ cells under similar conditions. Background
fluorescence was indicated by staining cells with IgG conjugated to FITC as
isotype control (Fig. lB). These results indicate that a population of Lin- cells
exist in human CB that do not produce intracellular or cell surface CD34.
Heterogeneity within human Lin-CD34+ cells is well documented and
further subdivision for the most p~ ilive cells is typically based on the cell
surface markers CD38, c-kit, Thy- 1 and HLA-DR. The expression of these
markers on both the Lin-CD34+ and Lin-CD34- cells was compared (Fig. lC).
Lin-CD34- cells displayed a bi-modal distribution of CD38 clearly dividing the
population into two fractions in contrast to the high proportion of Lin-CD34+
cells that express CD38 (Fig. lC). Cell surface expression of c-kit was similar
between that two populations, while the Lin-CD34- cells are almost exclusively
Thy-l- and HLA-DR- (Fig. lC). Both the absence of HLA-DR expression and
the presence of Thy-l have been proposed to define more plilllilive
subfractions within the Lin-CD34+ population. Therefore, the Lin-CD34-
population derived from CB is a distinct population which differs not only in
CA 02219869 1997-10-31
CD34 expression from plilllilive Lin-CD34+ cells but also in phenotypicheterogeneity based on additional markers associated with stem cells.
Example 2- Cell Immunostaining
Cord blood cells purified by flow cytometry as done previously (11,12)
were cytospun onto slides, permeabilized, and incubated in a BSA solution. The
results are shown in Figure lB. (Column I) Lin-CD34+CD38- cells were
stained with isotype control antibody conjugated to FITC (Becton Dickinson)
and countered stained with DNA binding DAPI as a control for non-specific
background flourescence. (Column II) Lin-CD34+CD38- cells and (Column
III) Lin-CD34- cells were stained with CD34 monoclonal antibodies followed
by DAPI counter stain. All slides were examined using a fluorescent
microscope ~ltili~ing the al~propliate filters for DAPI to detect the nucleus ofcells and FITC for the presence of CD34 protein.
Both Lin-CD34- and Lin-CD34+ purified cells were stained with
monoclonal antibodies conjugated to flourochromes for CD38 (Becton
Dickinson, BD), c-kit (BD), Thy-l (Coulter) and HLA-DR (BD). Stained
populations were then washed and analyzed using standard flow cytometric
techniques (J. Exp Med, PNAS) followed by the display of histograms using
the Cell Quest software program (BD) (n=3).
Example 3 - Cell Engraftment
Purified cell populations at the indicated dose were transplanted by tail
vein injection into sublethally irradiated mice (375 cGy using a 137Cs g-
irradiator) according to a standard protocol as previously described (16,17).
Mice were sacrificed 8 to 12 weeks post transplant and the BM from the
femurs, tibiae and iliac crests of each mouse were flushed into IMDM
contaillillg 10% FCS. Mouse BM cells was analyzed using FACS analysis and
by southern analysis using genomic DNA extracted by standard protocols in
which the level of human cell engraftment was determined by comparing the
18
CA 02219869 1997-10-31
characteristic 2.7 kb band with those of hl-m~n:mouse DNA mixtures as
controls (limit of detection 0.05% human DNA) (16, 17). The results are
shown in Figures 2 and 7.
Example 4 - Determin~tion of Hematopoietic Progenitor Activity of Lin-CD34-
Lin-CD34+CD38- and Lin-CD34-CD38+
Highly purified cells were plated in clonogenic methlycellulose assays
and seeded on MS-5 stroma in order to qua~ ale the CFC and LTC-IC content,
respectively. Clonogenic capacity of Lin-CD34- cells was extremely low in
comparison to Lin-CD34+CD38- cells (250 CFC vs. 8.9 CFC per 800 cells)
(Table I). As many as 10,000 cells needed to be seeded on MS-5 stroma to
detect a single LTC-IC within the Lin-CD34- cell fraction, while further
purification demonstrated that detection of LTC-IC in the Lin-CD34-CD38-
fraction required seeding of at least 2000 cells. By contrast, as few as 10 Lin-CD34+CD38- cells contain an LTC-IC. The Lin-CD34-CD38+ cells were
devoid of LTC-IC activity (limit of detection at 10,000 cells) but contained a
much higher capacity to form CFC specifically col-llllilled to the erythroid
lineage (Table I). The low efficiency to produce myeloid and erythroid
colllll-illed progenitors as well as the more ~l;lllilive LTC-IC is similar to
observations made with murine Lin-CD34- cells in the same assay systems and
suggest functional similarities may exist between the Lin-CD34- cells from
these two species (8, 10).
Example 5 - Multilineage Differentiation of Human lin-CD34- cells in
NOD/SCID mice after ex vivo Culture
A representative mouse was transplanted with 50,000 expanded lin-
CD34-CB cells after 2 days of ex-vivo culture in the presence of SF medium
supplemented with 5% FCS. Mouse BM was extracted 10 weeks after
transplant and analyzed by multiparameter flow cytometry (11, 12). The results
are shown in Figure 8.
19
CA 02219869 1997-10-31
CD34- Cell Culture with Growth Factors
Lin-CD34- cells were incubated in 50 ml of SF condition consisting of
IMDM supplemented with 1% BSA (Stem Cell Technologies),5 mg/ml of
human insulin (Humulin R from Eli Lilly and Co.),100 mg/ml of human
transferrin (Gibco, BRL),10 mg/ml of low density lipoproteins (Sigma
Chemical Co.), 10-4 M Beta-mercaptoethanol and growth factors (GF). GF
cocktail was used at final concentrations of 300 ng/ml of SCF (Amgen) and Flt-
3 (Immunex), 50 ng/ml of G-CSF (Amgen),10 ng/ml of IL-3 (Amgen) and IL-
6 (Amgen). 25% of condition media obtained from a fresh umbilical vein
endothelial cell culture in a low percentage of serum ( 10%) and passaged four
times, was added in some wells. Cells were cultured in flat bottomed
suspension wells of 96-well plates (Nunc), incubated for 2 and 4 days at 37~C
and 5% CO2 and 50 ml of fresh GF cocktail was added to each well every other
day.
Example 6 - Effect of ex vivo culture on the number of clonogenic pro~enitors
present in the lin- CD34-~ lin-CD34-CD38- and lin-CD34-CD38+ cell fractions
An aliquot of 800 to 2,500 lin-CD34-, lin- CD34-CD38- or lin-CD34-
CD38+ cells were plated in clonogenic progenitor assays under standard
conditions at the initiation of ex vivo cultures (day 0). Cells present after 4 days
of culture in the presence of SF or SF supplemented with 5% FCS or 25% of
HWEC-CM were plated in the same conditions. The number of CFC/800
input cells were estimated (mean ~ SEM; n-number of experiment). The
results are seen in Table II.
Example 7 - The effect of liquid culture on the development and potential
differentiation of lin-CD34-CD38- cells
Individual wells were analyzed by flow cytometry after 2 and 4 days of
culture in various conditions (Fig.6). Lin-CD34-CD38- cells seeded in SF
CA 02219869 1997-10-31
media began to express CD34 which could be enhanced with the addition of
serum (Fig.6). In contrast, the majority of cells obtained after 2 or 4 days of
culture in the presence of 25% HWEC conditioned medium still maintained
the same phenotype. The acquisition of CD34 demonstrates the differentiation
capacity of Lin-CD34-CD38- these cells in-vitro. The production of
CD34+CD38- cells suggests that lin-CD34-CD38- cells precede CD34+CD38-
population in the hierarchy of human hematopoiesis.
Example 8- Cytokine stimulation of Lin-CD34-CD38- Repopulating ActivityThe percentage of engrafted mice were compared before and after ex-
vivo culture of Lin- CD34-, lin- CD34-CD38- and lin-CD34-CD38+ fractions.
Purified fractions from 43 CB and 3 BM samples were cultured for 2 and 4
days in SF, SF supplemented with 5%FCS or 25% HWEC-CM and
transplanted at various doses into 136 recipient NOD/SCID mice (Fig. 7). Mice
transplanted with cultured lin-CD34-CD38+ cells were not engrafted consistent
with the inability of this fraction to engraft at day 0 (limit of detection <
0.05%). Transplanted Lin-CD34- cells we capable of engrafting mice with as
few as 8, 700 Lin-CD34- CB cells cultured for 4 days in SF conditions. A total
of 13 out 29 mice transplanted with cultured Lin-CD34- cells (cell dose ranging
from 8, 700 to 55, 000) were engrafted, indicating that ex-vivo culture
significantly increases the frequency of SRC (1 in 30,000 versus 1 in 125,000 atday 0). The increase in SRC activity could be explained by the acquisition of
the CD34 antigen due to a differentiation response of a subfraction of lin-
CD34-CD38- cells (Fig.6). However, similar increases in SRC was observed
after stimulation of Lin-CD34-CD38- cells for 4 days in the presence of
SF+25% HWEC-CM which does not cause differentiation into CD34+CD38-
cells. As many as 35 % of the mice transplanted with cultured lin-CD34-CD38-
cells engraft compared to 9% at day 0 ( Fig. l). In contrast,5,000 and 10,000
lin-CD34+CD38- CB cells CO~ g 10-20 SRC cultured for 4 days in the
presence of serum or the addition of 25% HWEC-CM were unable to engraft
21
CA 02219869 1997-10-31
NOD/SCID confirming the absence of collt~ ting SRC from this fraction in
Lin-CD34-CD38- repopulating cells (data not shown). The fact that
repopulating cells within lin-CD34+CD38- and lin-CD34-CD38- fractions
respond differently to the presence of 25%HWEC-CM, and that no phenotypic
change of lin-CD34-CD38- cells occurred after 4 days of culture suggests that
elements other than the acquisition of CD34 antigen must play a role in the
increase of repopulating activity of Lin-CD34-CD38- cells.
CA 02219869 1997-10-31
Table OneHematopoietic Progenitor Activity of Lin-CD34-~ Lin-CD34-CD38-
and Lin-CD34-CD38+
Phenotype of purified CFC/800 cellsFrequency of
population LTC-IC
CD34-Lin- 8.9+3 <1/10,000 (n=2)
(n=4) 5,501_1640(n=2)
CD34-Lin-CD38- 2.7_2 ~1/2,000
(n=36) (n=4)
CD34-Lin-CD38+ 77+31 <1/10,000 (n=10)
(n=4)
CD34+CD38- 250_46 >1/10
(n=4) (n=4)
CA 02219869 1997-10-31
Table Two
Effect of ex vivo culture on the number of clonogenic progenitors
present in the lin- CD34-, lin-CD34-CD38- and lin-CD34-CD38+ cell fractions.
Phenotype of purified Da~l 0 After in vitro stimulation
population
CD34-Lin- 8.9 _3 Serum Free 10 _0.9 (n=4)
(n=4) 5% Serum 76 _21 (n=2)
CD34-Lin-CD38- 2.7_ 2 5% Serum 6.4 _4.5 (n=7)
(n=36) Serum Free 1.2 _1 (n=9)
25% HWEC 0.5% _0.5 (n=7)
CD34-Lin-CD38+ 77 _31 Serum Free 20 _10 (n=4)
(n=10) 5% Serum 134_ 41 (n=4)
*CFC/800 Input Cells
24
CA 02219869 1997-10-31
References
l.Till, J.E., McCulloch, E.A. 1980. Hemopoietic stem cell
differentiation. Biochim.Biophys. Acta. 605(4):431-59.
2. Ogawa, M. 1993. Differentiation and proliferation of hematopoietic
stem cells . Blood 81: 2844-2853.
3. Morrison, S., Weissman, I.L.1994. The long- term repopulating subset
of hematopoietic stem cells is deterministic and isolable by phenotype.
Immunity 1: 661-667, 1994.
4. Krause, D.S., Fackler, M.J., Civin, C.I., May, W.S. 1996. CD34:
structure, biology, and clinical utility. Blood 87: 1-12.
5. Spangrude, G.I., Heimfeld, S., Weissman, I.L. 1988. Purification and
characterization of mouse hematopoietic stem cells. Science 241: 58-62.
6. Spangrude, G.J., Brooks, D.M. 1992. Phenotypic analysis of mouse
hematopoietic stem cells shows a Thy-l negative subset. Blood 80: 1957-1963.
7. Craig, W., Kay, R., Cutler, R.L., Lansdorp, P.M. 1993. Expression of
Thy-l on human hematopoietic progenitor cells. J.Exp.Med. 177: 1331-1342.
8. Goodell, M.A., Brose, K., Paradis, G., Conner, A.S., and Mulligan,
R.C. 1996. Isolation and functional properties of murine stem cells that are
replicating in vivo. J.Exp.Med 183: 1797-1806.
9. Jones, R.J., Collector, M.I., Barber, J.P., Vala, M.S., Fackler, M.J.,
May, S.W., Griffin, C.A., Hawkins, A.L., Zehnbauer, B.A., Hilton, H., Colvin,
O.M., Sharkis, S.J. 1996. Characterization of mouse lymphohematopoietic stem
cells lacking spleen colony-forming activity. Blood 88: 487-491.
10. Osawa, M., H~n~d~ K.I., H~m~d~ H., Nakauchi, H. 1996 Long-
term lymphohematopoietic reconstitution by a single CD34-low/ negative
hematopoietic stem cell. Science 273: 242-245.
11. Bhatia, M., Bonnet, D., Kapp, U., Wang, J.C.Y., Murdoch, B., Dick,
J.E. 1997. Quantitative analysis reveals expansion of human hematopoietic
cells after short-term ex-vivo culture. J.Exp.Med. 186: 619-624.
-
CA 02219869 1997-10-31
12. Bhatia, M., Wang J.C.W.,, Kapp, U., Bonnet, D., Dick, J.E . 1997.
Purification of primitive human hematopoietic cells capable of repopulating
immunodeficient mice. Proc Natl Acad Sci USA. 94: 5320-5325.
13. Lapidot, T., F. Pflumio, M. Doedens, B. Murdoch, D. E. Williams
and J. E. Dick. 1992. Cytokine stimulation of multilineage hematopoiesis from
imm~ re human cells engrafted in scid mice. Science. 255: 1137-1141.
14. Dick, J. E. 1996. Normal and leukemic human stem cells assayed in
SCID mice. Sem. Immunol. 8: 197-206.
15. Larochelle, A., J. Vormoor, H. Hanenberg, J. C. Y. Wang, M.
Bhatia, Lapidot.T, T. Moritz, B. Murdoch, L. X. Xiao, I. Kato, D. A. Williams
and J. Dick. 1996. Identification of pli~nilive hematopoietic cells capable of
repopulating NOD/SCID mice: implications for gene therapy. Nature Med. 2:
1329-1337.
16. Petzer, A. L., P. W. Zandstra, J. M. Piret and C. J. Eaves. 1996.
Differential cytokine effects on plilllilive (CD34+CD38-) human hematopoietic
cells: novel responses to Flt3-ligand and thrombopoietin. J .Exp. Med. 183:
2551-2558.
17. Conneally, E., Cashman, J., Petzer, A., Eaves, C.J. 1997. Proc Natl
Acad Sci USA 94: 9836-9841.
18. Hao, Q., F. T. Thiemann, D. Peterson, E. M. Smogorzewska and G.
M. Crooks. 1996. Extended long-term culture reveals a highly quiescent and
plilllilive human hematopoietic progenitor population. Blood. 88: 3306-3313.
26
