Note: Descriptions are shown in the official language in which they were submitted.
CA 02358326 2001-10-O1
-1 -
B8P File No. 7771-74
TITLE: METHOD FOR SEPARATING CELLS USING DISCONTINUOUS
DENSITY GRADIENT CENTRIFUGATION
FIELD OF THE INVENTION
The present invention relates to methods for separating cells. In
particular the invention relates to methods for separating cells using
discontinuous density gradient centrifugation.
BACKGROUND OF THE INVENTION
In many applications it is desirable to enrich or, alternatively, deplete
certain cell populations in a biological sample. The fields of hematology,
immunology and oncology rely on samples of peripheral blood and cell
suspensions from related tissues such as bone marrow, spleen, thymus and
fetal liver. The separation of specific cell types from these heterogeneous
samples is key to research in these fields, diagnostics and therapy for
certain
malignancies and immune/hematopoietic disorders.
Purified populations of immune cells such as T cells and antigen
presenting cells are necessary for the study of immune function and are used
in immunotherapy. Investigation of the cellular, molecular and biochemical
processes require analysis of certain cell types in isolation. Numerous
techniques have been used to isolate T cell subsets, B cells, basophils, NK
cells and dendritic cells.
The isolation of hematopoietic stem cells has also been an area of
great interest. Pure populations of stem cells will facilitate studies of
hematopoiesis and transplantation of hematopoietic cells from peripheral
blood and/or bone marrow is increasingly used in combination with high-dose
chemo- and/or radiotherapy for the treatment of a variety of disorders
including malignant, nonmalignant and genetic disorders. Very few cells in
such transplants are capable of long-term hematopoietic reconstitution, and
thus there is a strong stimulus to develop techniques for purification of
hematopoietic stem cells. Furthermore, serious complications and indeed the
success of a transplant procedure is to a large degree dependent on the
effectiveness of the procedures that are used for the removal of cells in the
CA 02358326 2001-10-O1
-2 -
transplant that pose a risk to the transplant recipient. Such cells include T
lymphocytes that are responsible for graft versus host disease (GVHD) in
allogenic grafts, and tumor cells in autologous transplants that may cause
recurrence of the malignant growth. It is also important to debulk the graft
by
removing unnecessary cells and thus reducing the volume of cyropreservant
to be infused.
In certain instances it is desirable to remove or deplete tumor cells from
a biological sample, for example in bone marrow transplants. Epithelial
cancers of the bronchi, mammary ducts and the gastrointestinal and
urogenital tracts represent a major type of solid tumors seen today.
Micrometastatic tumor cell migration is thought to be an important prognostic
factor for patients with epithelial cancer (Braun et al., 2000; Vaughan et
al.,
1990). The ability to detect such metastatic cells is limited by the
effectiveness of tissue or fluid sampling and the sensitivity of tumor
detection
methods. A technique to enrich circulating epithelial tumor cells in blood
samples would increase the ability to detect metastatic disease and facilitate
the study of such rare cells and the determination of the biological changes
which enable spread of disease.
Hematopoietic cells and immune cells have been separated on the
basis of physical characteristics such as density and on the basis of
susceptibility to certain pharmacological agents which kill cycling cells. The
advent of monoclonal antibodies against cell surface antigens has greatly
expanded the potential to distinguish and separate distinct cell types. There
are two basic approaches to separating cell populations from blood and
related cell suspensions using monoclonal antibodies. They differ in whether
it is the desired or undesired cells which are distinguished/labeled with the
antibody(s).
In positive selection techniques the desired cells are labeled with
antibodies and removed from the remaining unlabeled/unwanted cells. In
negative selection, the unwanted cells are labeled and removed.
Antibody/complement treatment and the use of immunotoxins are negative
selection techniques, but FACS sorting and most batch wise
CA 02358326 2001-10-O1
-3 -
immunoadsorption techniques can be adapted to both positive and negative
selection. In immunoadsorption techniques cells are selected with
monoclonal antibodies and preferentially bound to a surface which can be
removed from the remainder of the cells e.g. column of beads, flasks,
magnetic particles. Immunoadsorption techniques have won favour clinically
and in research because they maintain the high specificity of targeting cells
with monoclonal antibodies, but unlike FACSorting, they can be scaled up to
deal directly with the large numbers of cells in a clinical harvest and they
avoid the dangers of using cytotoxic reagents such as immunotoxins, and
complement. They do however, require the use of a "device" or cell
separation surface such as a column of beads, panning flask or magnet.
Current techniques for the isolation of hematopoietic stem cells,
immune cells and circulating epithelial tumor cells all involve an initial
step to
remove red cells then antibody mediated adherence to a device or artificial
particle. (First et al., 1988; de Wynter et al., 1975; Shpall et al., 1994;
Thomas et al., 1994; Miltenyi Biotec Inc., Gladbach, Germany) In the case of
positive selection there is yet another step; removal of the cells from the
device or particle. All these multiple steps require time and incur cell loss.
Discontinuous density gradient centrifugation is commonly used to
isolate peripheral blood mononuclear cells from granulocytes and
erythrocytes. Ficoll-Paque~ (Amersham Pharmacia Biotech AB, Uppsala
Sweden) is the most popular density separation solution used for this
application. In a Ficoll density separation whole blood is layered over
Ficoll,
and then centrifuged. The erythrocytes and granulocytes settle to the cell
pellet and the mononuclear cells remain at the Ficoll plasma interface. The
success of this technique relies on the difference in density between
mononuclear cells and granulocytes/erythrocytes and the choice of the
density separation medium (DSM). During centrifugation, cells that are more
dense than the density separation medium settle through the DSM forming a
pellet at the bottom of the tube, while cells that are less dense than the DSM
collect at the interface between the DSM and the cell suspension medium
(e.g. plasma in the case of peripheral blood, cell culture medium in the case
of
CA 02358326 2001-10-O1
-4 -
cultured cells or dissociated tissue cells). Multiple layers of DSM having
different densities can be used to divide the cells into multiple fractions.
The
same effect can be achieved by centrifuging cells in a medium with a
continuous density gradient and then collecting the cells from adjacent strata
in the gradient. Sedimentation rate can also be used to separate cells of
different density, but the separation is influenced not only by density but
also
the viscosity of the suspension and the cell size.
All density separation techniques have the same basic limitation; they
can not separate subpopulations of cells with overlapping density
distributions. They do not offer the high cell specificity offered by antibody
mediated techniques. Dense particles have been targeted to cells using
monoclonal antibodies with affinity to cells surface antigens and used in
discontinuous or continuous density gradient centrifugation to separate cell
populations with similar densities [Bildirici and Rickwood (2001), Bildirici
and
Rickwood (2000), Patel and Rickwood (1995) and Patel et al (1993), US
patent 5,840,502 and StemCell Technologies, Supplement to 1999/2000
Catalogue supplement]. The disadvantage of continuous density gradient
separations for cell enrichment using dense particles is that there is no
clear
delineation between the desired particle-free cells and the undesired particle-
bound cells. Several patents (US 5,840,502, US 5,648,223, US 5,646,004 and
US 5,474,687) describe the use of dense particles to selectively deplete
targeted cell types by discontinuous density separation. These patents state
that the optimum density of the DSM for dense particle separation is within
~0.0005 to ~0.0002 g/cm3 of the density of the desired cell population.
SUMMARY OF THE INVENTION
The invention relates to the use of discontinuous density gradient
centrifugation to separate desired populations of cells from a mixed cell
mixture or solution. It has been found that the use of dense particles and a
density separation medium (DSM) with a density at least 0.001 g/cm3 higher
than the density of the desired cells offers more effective cell separation
through increased recovery of the desired cells without affecting the purity
of
CA 02358326 2001-10-O1
-5 -
the desired cells. This is contrary to current practice and the teachings of
US
Patent Nos. 5,840,502, US 5,648,223, US 5,646,004 and US 5,474,687.
Accordingly, the present invention provides a method for separating
desired cells from undesired cells in a sample comprising:
- linking dense particles to the undesired cells in the sample to
increase the effective density of the undesired cells;
- layering the sample over a density separation medium (DSM)
having a density at least about 0.001 glcm3 greater than the mean
density of the desired cells;
- centrifuging the sample and DSM; and
- recovering the desired cells from the interface between the DSM
and the sample.
Alternatively, the dense particles could be linked to the desired cells,
followed by isolation of the desired cells from the pellet at the bottom of
the
sample and cleavage of the dense particles from the cells using enzymatic or
chemical methods.
The invention includes all uses of the above-described methods as well
as kits to perform the methods of the invention.
Other features and advantages of the present invention will become
apparent from the following detailed description. It should be understood,
however, that the detailed description and the specific examples while
indicating preferred embodiments of the invention are given by way of
illustration only, since various changes and modifications within the spirit
and
scope of the invention will become apparent to those skilled in the art from
this detailed description.
DETAILED DESCRIPTION OF THE INVENTION
As hereinbefore stated, it has been found that the use of dense
particles and a density separation medium (DSM) with a density at least 0.001
g/cm3 higher than the mean density of the desired cells offers more effective
cell separation when using discontinuous density gradient centrifugation
methods.
CA 02358326 2001-10-O1
-6 -
The method of the invention may be applied using negative or positive
selection techniques. Preferably, negative selection is used, therefore the
present invention provides a method for separating desired cells from
undesired cells in a sample comprising:
- linking dense particles to the undesired cells in the sample to
increase the effective density of the undesired cells;
- layering the sample over a density separation medium (DSM)
having a density at least about 0.001 glcm3 greater than the mean
density of the desired cells;
- centrifuging the sample and DSM; and
- recovering the desired cells from the interface between the DSM
and the sample.
Alternatively, positive selection may be used therefore the present
invention provides a method for separating desired cells from undesired cells
in a sample comprising:
- linking dense particles to the desired cells in the sample to increase
the effective density of the desired cells;
- layering the sample over a density separation medium (DSM)
having a density at least about 0.001 g/cm3 greater than the mean
density of the undesired cells;
- centrifuging the sample and DSM;
- recovering the desired cells linked to the dense particles from the
bottom of the sample; and
- cleaving the dense particles from the desired cells.
Cleavage of the dense particles from the desired cells may be carried
out using enzymatic or chemical methods well known in the art. For example,
using proteolytic enzymes such, as papain.
The sample can be any sample from which one wishes to separate
particular or desired cells. As such, the sample will generally contain a
mixture of desired cells and undesired cells. The sample can be obtained
from both in vivo or in vitro sources. Examples of sources include, but at not
limited to, peripheral blood, bone marrow, spleen, thymus and fetal liver.
CA 02358326 2001-10-O1
_7 _
Specific cell types that may be isolated using the method of the invention
include, but are not limited to, T cells, B cells, basophils, NK cells,
dendritic
cells, monocytes, macrophages, megakaryocytes, platelets, eosinophils,
neutrophils, hematopoietic stem cells, mesenchymal stem cells, endothelial
cells, epithelial cells, fibroblasts and tumour cells.
Density Separation Medium
A variety of commercially available gradient materials may be used in
the method of the invention, including, but not limited to Ficoll-Paque~;
Lymphoprep~; any sugar solution, e.g. sucrose; dextran; any protein solution,
e.g. bovine serum albumin (BSA); iodinated low molecular weight compounds
such as Metrizamide and heavy salts, e.g. cesium chloride. Preferably, the
density separation medium (DSM) is prepared by mixing hetastarch, iodixanol
and water in different proportions such that the desired density and
osmolarity
is obtained.
The density of the DSM should be at least about 0.001 g/cm3, preferably
about 0.002 g/cm3, more preferably about 0.004 g/cm3, higher, than the mean
density of the density of the desired cells. The density of the DSM should
also be lower than the density of the dense particles used for the separation.
As used herein, the term "mean density of the desired cells" refers to the
density of the desired cells as determined using density gradient
centrifugation with a series of DSM having different densities and without
dense particles (for example, see Example 5 herein). The DSM is said to have
a density equal to the "mean density of the desired cells" if the number of
desired cells recovered at the interface between the DSM and the cell
suspension medium is equal to the number of desired cells recovered in the
pellet after performing density gradient separation (or if 50% of the
initially
present desired cells are recovered at the interface when recovery in the
pellet cannot be reliably determined).
The density of the DSM should also be lower than the effective density
of the undesired cells when linked to dense particles such that most of the
undesired cells settle to the pellet during density gradient separation. The
effective density of the undesired cells will depend on the size and density
of
CA 02358326 2001-10-O1
_$ _
the cells and the size, density and number of particles linked to the cells.
For
any cell linked to dense particles, the effective density can be defined as
the
product of the cell volume (V~) and cell density (p~) plus the product of the
volume of the particles (VP) linked to the cell and the density of the
particles
(pp) all divided by the total volume of the cell and the particles linked to
the
cell:
y°Pc + VPPP
1o v~+vP
For example, an undesired cell with a diameter of 10 Nm and a density
of 1.0 g/cm3 linked to one particle with a diameter of 10 Nm and a density of
2.0 g/cm3 would have an effective density of 1.5 g/cm3. The same undesired
cell linked to 9 particles with a diameter of 10 pm and a density of 2.0 g/cm3
would have an effective density of 1.9 g/cm3.
Alternatively, when using positive selection techniques, a person
having skill in this art would understand that the density of the DSM should
be
at least about 0.001 g/cm3, preferably about 0.002 g/cm3, more preferably
about 0.004 g/cm3, higher, than the mean density of the density of the
undesired cells.
The osmolarity of the DSM is preferably approximately the same as the
osmolarity of the sample. The term "approximately the same" means that the
osmolarity of the DSM is in the range of about plus or minus 10% of the
osmolarity of the sample. Most preferably, the osmolarity of the DSM is in the
range of about 270 to about 300 mOsm.
The method of the invention may be carried out in any suitable
centrifuge tube or syringe.
Dense Particles
The use of particles to adjust the density of the undesired or desired
cells during density gradient cell separations is described in US 5,840,502,
the contents of which are incorporated herein by reference. The method of
CA 02358326 2001-10-O1
_g _
the present invention may be applied to negative selection techniques (dense
particles linked to undesired cells - preferred) or positive selection
techniques
(dense particles linked to desired cells).
In preferred embodiments of the present invention, the cells and dense
particles may be linked through various types of binding including, but
limited
to, drug-drug receptor, antibody-antigen, hormone-hormone receptor, growth
factor-growth factor receptor, carbohydrate-lectin, nucleic acid
sequence-complementary nucleic acid sequence, enzyme-cofactor or
enzyme-inhibitor binding. Preferably, the cells to be linked to the dense
particles are defined by specific surface proteins and the dense particles are
linked to these cells by antibodies specific for the cell surface proteins.
In further embodiments of the present invention, the dense particles
may be selected from the group consisting of red blood cells, silica
particles,
metal particles, polymer particles, glass particles. Red blood cells may be
linked to the undesired or desired cells using tetrameric antibody complexes
as described in Example 2 herein. Other particles may be linked to the
undesired or desired cells by coating the particles with, for example,
antibodies or dextran, as described in Example 3 herein.
Kits
Density separation media, cell specific targeting agents andlor dense
particles for the separation of specific cell types may be prepared and
packaged in convenient kits, packaged into suitable containers. For example,
for applications that use red blood cells as dense particles to isolate cells
from
whole peripheral blood, as in Example 2, the reagents may include one or
more aliquots of a density separation medium having a density at least about
0.001 g/cm3 higher than the mean density of the desired cells, and one or
more aliquots of tetrameric antibody complex cocktails that link the undesired
cells to the red blood cells. Alternatively, the reagents may include one or
more aliquots of a density separation medium having a density at least about
0.001 g/cm3 higher than the mean density of the undesired cells, and one or
more aliquots of a tetrameric antibody complex that links the desired cells to
the red blood cells.
CA 02358326 2001-10-O1
-10 -
As a second example, for isolating cells from a sample consisting more
generally of nucleated cells, the reagents may include one or more aliquots of
a density separation medium having a density at least about 0.001 g/cm3
higher than the mean density of the desired cells, and either one or more
aliquots of antibody coated dense particles that target the undesired cells,
or
one or more aliquots of an antibody cocktail that bind to the undesired cells
and one or more aliquots of dense particles that then bind to the cells
through
interactions with the antibodies in the antibody cocktail. Alternatively, the
reagents may include one or more aliquots of a density separation medium
having a density at least about 0.001 g/cm3 higher than the mean density of
the undesired cells, and either one or more aliquots of antibody coated dense
particles that target the desired cells, or one or more aliquots of an
antibody
that binds to the desired cells and one or more aliquots of dense particles
that
then bind to the desired cells through interactions with the antibody, and/or
one or more aliquots of a reagent to cleave the particles from the cells.
With particular regard to density separation media packaged in "kit"
form, it is preferred that aliquots of the media be packaged in separate
containers, with each container including a sufficient quantity of reagent for
at
least one assay to be conducted. A preferred kit is typically provided as an
enclosure (package) comprising one or more containers for the within-
described reagents.
Printed instructions providing guidance in the use of the packaged
reagents) may also be included, in various preferred embodiments. The term
"instructions" or "instructions for use" typically includes a tangible
expression
describing the reagent concentration or at least one assay method parameter,
such as the relative amounts of reagent and sample to be admixed,
maintenance time periods for reagent/sample admixtures, temperature, buffer
conditions, and the like.
The following non-limiting examples are illustrative of the present
invention:
EXAMPLES
Example 1 - Preparation of density separation media (DSM)
CA 02358326 2001-10-O1
-11 -
Density separation media (DSM) were prepared at different densities
by mixing hetastarch, iodixanol and water in different proportions such that
the
desired density was obtained and the osmolarity was between 270 and 300
mOsm. The hetastarch and iodixanol components of this DSM serve the
same function as the polysaccharide and metrizoate components respectively
in both Ficoll-Paque~ and Lymphoprep~.
Example 2 - Method to Isolate Cells from Whole Human Peripheral
Blood Using the Method of the Invention and Red Cells as Dense
Particles
Preparation of Tetramers
In order to prepare a tetrameric antibody complex for use in the method
of the present invention, the following protocol may be used: (a) take 1 mg of
antibody specific for cells to be rosetted (e.g. anti-CD2, CD3, CD4, CDB,
CD14, CD16, CD19 etc.); (b) add 3 mg anti-Glycophorin A antibody (against
red blood cells); mix well (c) then add 4.0 mg of P9 antibody or 2.72 mg of
the
P9 F(ab')2 antibody fragment. Incubate overnight at 37°C. The P9
antibody
binds the Fc portion of the antibodies added in steps (a) and (b) resulting in
a
tetrameric antibody complex. For more information on the preparation of
tetramers see U.S. Patent No. 4,868,109 to Lansdorp, which is incorporated
herein by reference. Tetrameric antibody complexes incorporating different
antibodies to antigens expressed on nucleated cells are prepared separately
and then mixed.
The antibody compositions are made by combining various tetrameric
antibody complexes depending on which cells one wishes to deplete. The
concentration of the various tetrameric antibody complexes varies: typically
antibodies to antigens expressed on nucleated cells are at 10-30 ~g/mL in
tetrameric complexes. The composition is then diluted 1/10 into the cells so
the final concentrations of each anti nucleated cell antibody in the cell
suspensions is 1.0-3.0 ~g/mL.
Separation method
A negative selection protocol for immunorosetting cells from whole
peripheral blood using Ficoll Hypaque is set out below.
CA 02358326 2001-10-O1
-12 -
1. Add 100pL antibody composition per mL of whole peripheral blood.
2. Incubate 20 minutes at room temperature.
3. Dilute sample with an equal volume of phosphate buffered saline (PBS)
+ 2% fetal calf serum (FCS) and mix gently.
4. Layer the diluted sample on top of Ficoll Hypaque or layer the Ficoll
underneath the diluted sample.
5. Centrifuge for 20 minutes at 1200 x g, room temperature, with the
brake off.
6. Remove the enriched cells from the Ficoll:plasma interface.
7. Wash enriched cells with 5-10x volume of PBS + 2% FBS.
Note: For enrichment of monocytes and other adherent cells, add 1 mM
EDTA to the sample of whole blood and to all wash/dilution solutions.
Example 3 - Preparation of surface modified silica particles
Dense silica particles were coated with antibodies or dextran. The
antibodies and dextran were present during the coating step at sufficient
concentration to provide at least one molecule monolayer coverage of the
particles.
Example 4 - Method to isolate cells from human peripheral blood
mononuclear cells, murine spleen cells or murine bone marrow cells by
density separation using the method of the invention and silica particles
as dense particles.
Single cell suspensions were obtained from murine spleen or bone
marrow using standard procedures. Human mononuclear cell suspensions
were prepared by Ficoll-Paque~ density separation. Antibodies that bind to
specific cell surface proteins on unwanted cells were added to the cell
suspension at a final concentration of between 1.0 and 3.0 pg/mL and
incubated for at least 5 min with the cells. For murine cell enrichments, rat
anti-mouse antibodies from various suppliers were used. For example to
enrich for murine CD4+ cells, antibodies targeting CD11b, CD45R, CDB,
erythroid cells (TER119) and myeloid differentiation antigen (Gr-1) were
added. For human cell enrichments, bi-specific antibody complexes were
used. These antibody complexes are formed by mixing one part human cell
CA 02358326 2001-10-O1
-13
specific murine IgG1 antibody (eg. anti-human CD2, CD3, CD4, CDB, CD14,
CD16 etc.) with two parts murine IgG1 anti-dextran antibody and then adding
three parts P9 F(ab')2 antibody fragment. For example, to deplete CD3+ cells,
anti-CD3 tetramers were added. More information on the preparation of this
antibody complex is in US Pat. No. 4,868,109.
The cells were washed and then incubated with coated silica particles,
prepared as in Example 3, that bind to the antibody labeled cells. For murine
cell isolations, GAM-coated particles were used. For human cell isolations,
dextran coated particles were used. The cell-particle mixture was then layered
over different DSM and centrifuged for 20 min at 1200 x g with no brake. The
interface cells containing the desired cell population were collected and the
purity of the desired cell population or the depletion of undesired cell
population was determined by flow cytometry using standard procedures.
Example 5 - Evaluation of DSM for the separation of murine spleen cells
without dense particles
Single cell suspensions were obtained from mouse spleens using
standard procedures. A total of 2 x 10' cells was diluted in 2 mL of the
appropriate buffer and used in each of the test DSM. Three (3) mL of the
various density media were aliquoted in separate 15 mL conical
polypropylene tubes. The density separation media and their respective
density and osmolarity were Ficoll-Paque Plus from Pharmacia (1.077, g/cm3,
300-310 mOsm) and three density separation media (DSM) prepared as
described in Example 1 with densities of 1.077, 1.081 and 1.085 glcm3 and
Osm=290 mOsm. The diluted cell suspensions were layered on top of each of
the density media. The samples were centrifuged in a swinging bucket
centrifuge at 1200 x g for 10 minutes at room temperature with the centrifuge
brake off. The original spleen cell suspension, light density cells (collected
from the buffer:density medium interface) and pellet cells were analyzed by
flow cytometry.
The results in Table 1 show that the recovery of lymphocytes and
granulocytes increases with increasing density of the DSM up to 1.085 g/cm3.
The recovery with Ficoll-Paque~ is lower than with the DSM prepared with the
CA 02358326 2001-10-O1
-14 -
same density likely because it has a lower osmolarity than Ficoll-Paque~. A
DSM density of 1.081 g/cm3 is ideal for depleting RBC while maintaining a
high lymphocyte recovery according to these data. The data in Table 1 also
show that the average lymphocyte and granulocyte density is between 1.077
and 1.081 g/cm3. According to the prior art a DSM with a density in this range
is preferred for dense particle assisted density separations. Further examples
will show that the preferred DSM density is higher than the average
lymphocyte or granulocyte density. Although the data in Table 1 relate
specifically to mouse spleen cells, this type of analysis can be applied to
any
cell population to determine the recovery and purity as a function of DSM
density.
Example 6 -Effect of DSM density on enrichment of desired murine
spleen and bone marrow cells using dense particles to deplete
unwanted cells during density separation
CD4+ cells were enriched from murine spleen suspensions following
the method of Example 4 using a cocktail of antibodies targeting CD11 b,
CD45R, CDB, erythroid cells (TER119) and myeloid differentiation antigen
(Gr-1 ). CD8+ cells were similarly enriched using a cocktail of antibodies
targeting CD11b, CD45R, CD4, erythroid cells (TER119) and myeloid
differentiation antigen. Murine progenitors, defined as Sca1+ / lineage
negative (CD3, CD11b, CD45R and Gr-1 negative)-, were enriched from
murine bone marrow using a cocktail of antibodies targeting CDS, CD11b,
CD45R, erythroid cells (TER119), myeloid differentiation antigen and
neutrophils (7-4).
The results in Table 2 show that recovery of the desired cells increases
with increasing density of the DSM and that the range of useful densities is
well above 1.081 g/cm3 which was shown in Example 5 to be the upper limit
of the density of the desired lymphocytes. Thus, it is advantageous to use a
DSM with a higher density than the density of the desired cell for murine cell
separations.
CA 02358326 2001-10-O1
-15 -
Example 7 - Effect of DSM density on enrichment of desired human
peripheral blood cells by density separation using red blood cells as
dense particles
Human peripheral blood cells were separated using the method
described in Example 2. The T (CD3+) cell enrichment cocktail included
antibody complexes targeting CD16, CD19, CD36 and CD56. The CD4+ T
cell enrichment cocktail included antibody complexes targeting CDB, CD16,
CD19, CD36 and CD56. The CD8+ T cell enrichment cocktail included
antibody complexes targeting CD4, CD16, CD19, CD36 and CD56. The B
(CD19+) cell enrichment cocktail included antibody complexes targeting CD2,
CD3, CD16, CD36 and CD56. The NK (CD56+) cell enrichment cocktail
included antibody complexes targeting CD3, CD4, CD19, CD36 and CD66b.
Table 3 shows data obtained from two different samples, where each sample
was incubated with the given cocktail of antibody complexes, diluted and then
layered in 3 equal volume aliquots over each of Ficoll-Paque~ and two DSM
prepared with densities of 1.081 and 1.085 g/cm3 and separated following
Example 1. The tabulated values are the average of the results for the n=3
separations for each DSM. The results in Table 3 clearly show that the
recovery of the desired cells is improved with increased DSM density and that
the purity of the desired cell population obtained is not adversely affected
up
to a density of 1.085 g/cm3. The results summarized in Table 4 are from the
separation of 3 samples where such sample was divided into 5 equal parts
and incubated with a cocktail of antibody complexes for the enrichment of the
5 desired cell types listed following the method of Example 2. Each of these
volumes was then diluted, split into two equal volumes and layered over either
Ficoll-Paque'"' or a DSM prepared with a density of 1.081 g/cm3. Table 4
shows that the recovery of desired cells is higher using a DSM with a higher
density than Ficoll-Paque~. The purity of the desired population is equivalent
or higher using DSM. The optimum DSM density for the isolation of T, B and
NK cells from whole blood in conjunction with dense particle enrichment is
thus higher than that of Ficoll-Paque~ which is close to the density of MNC
and is optimum for MNC isolation in the absence of dense particle
CA 02358326 2001-10-O1
-16
enrichment. This example shows that the preferred DSM density for density
separations of human lymphocyte subsets using red blood cells as dense
particles to deplete unwanted cells is substantially higher than the density
of
the desired lymphocytes.
Example 8 - Depletion of undesired human peripheral blood cells by
density separation using dextran coated silica particles
Frozen mononuclear cells were incubated with anti-CD3:anti-dextran
bi-specific antibody complex (as described in Example 4) at 1.0 ~g/mL for 20
min at room temperature, then washed to remove excess tetramer. Dextran
coated silica particles were then added at 2.5 pg/mL, incubated for 20 minutes
at room temperature and then diluted 4 times and layered over either Ficoll-
Paque~ or a DSM with a density of 1.081 g/cm3 in a centrifuge tube. The
tubes were centrifuged for 20 min at 1200x g with no brake and the cells at
the interface were recovered and washed 1 x. The CD3+ cell log depletion
was determined by taking the log of the number of cells labeled with sheep
anti-mouse (SAM)-FITC+ in the start sample divided by the number in the final
sample. SAM antibodies bind to cells that have been targeted by murine
antibodies including the anti-CD3: anti-dextran bi-specific antibodies.
Table 5 shows that the log depletion of CD3+ cells was equivalent for
both DSM but that the recovery of CD3- cells was higher using DM-L because
more cells were buoyant than in Ficoll-Paque~. The density of Ficoll-Paque"
is close to the mean lymphocyte density. This example shows that the
preferred density of the DSM for density separations using silica particle to
deplete unwanted cells is higher than the lymphocyte density to provide
improved recovery of CD3~ cells without affecting CD3+ cell depletion
efficiency.
While the present invention has been described with reference to what
are presently considered to be the preferred examples, it is to be understood
that the invention is not limited to the disclosed examples. To the contrary,
the
invention is intended to cover various modifications and equivalent
arrangements included within the spirit and scope of the appended claims.
CA 02358326 2001-10-O1
-17 -
All publications, patents and patent applications are herein
incorporated by reference in their entirety to the same extent as if each
individual publication, patent or patent application was specifically and
individually indicated to be incorporated by reference in its entirety.
CA 02358326 2001-10-O1
-18 -
Table 1 Recovery of murine red blood cells, lymphocytes and
granulocyte populations in the interface and pellet after density gradient
centrifugation of murine spleen cells over density media of various
densities
Interface Pellet
Density of Red LymphocytesGranulocytesRed LymphocytesGranulocytes
medium blood blood
(g/cm') cells % % cells
1.077 (Ficoll-Paque~)0.2 20 18 76 20 79
1.077 (in-house)0.2 48 50 92 50 42
1.081 (in-house)0.3 68 38 > 100 1.4 2
1.085 (in-house)16 86 80 69 1 1.13
1.077 (Ficoll-Paque~)< 0.1 5.3 10.4 > 100 83 96
1.077 (in-house)0.2 15 16 > 100 76 67
1.081 (in-house)0.3 101 > 100 > 100 5.9 12
1.085 (in-house)3.6 >100 > 100 25 1.9 3.2
Table 2 - Recovery of desired murine cells increases with increasing
DSM density for density separations using GAR-coated silica particles
DSM Densi ( icm')Start % Enriched % Recove
CD8- cells (n=2) (n=2)
1.0950 9.4 88, 86 __
49, 49
1.0900 9.4 90, 90 32, 29
Scam, lin- (n=2)
1.0900 1.6 10, 11 52, 39
1.0875 1.6 9.2, 9.4 47, 41
1.0800 1.6 8.6, 8.5 33, 37
CD4+ cells (n=2) (n=2)
l .0900 21 84. 81 34, 27
1.0875 21 88, 86 34, 34
1.0800 21 84, 84 29, 30
C D4+ ce n=1 n=1
I I s
1.0950 24 87 26
1.0900 24 89 23
CA 02358326 2001-10-O1
-19 -
Table 3 - Effect of DSM density on the enrichment of various human
lymphocyte populations from whole blood by density separation using
red blood cells as dense particles.
DSM % recovery desired
cell % purity
desired cells
T cell CD3+)
enrichment
n=3
Ficoll-Pa ue 1.07726 96
cm
DSM 1.081 icm 44 96
DSM 1.085 cm 55 96
B cell CD19+
enrichment
n=3
Ficoll-Pa ue 1.07762 86
/cm'
DSM 1.081 cm 96 88
DSM 1.085 o/cm 91 85
Table 4 - Comparison of Ficoll-Paque~ and DSM at 1.081 glcm3 in
density separations of various human lymphocyte populations using red
blood cells as dense particles (n=3 samples).
Desired CellPurity Recovery
Population Ficofl-PaqueDSM (1.081 Ficoll- DSM (1.081
g%cm ) Pa ue~' cm-)
T cell (CD3+)98% 98% 61% 71io
CD4+ T cell 94% 94% 67% 78~0
CD8+ T cell 85% 85% 42% 43io
B cell (CD19+)84% 90% 72% 87%
NK cell (CD~6+)82% 85% 24% 27%0
Table 5 - Depletion of human CD3+ T cells by dense particles for n=2
separations
Density separation medium °~o CD3~ Log depletion recovery CD3-
CD3T cells
Initial sample 39
DSM 1.081 g/cm' 4.3 1.6 34
Ficoll-Paque~ 9.9 1.4 26
Initial sample 39
DSM 1.091 g/cm3 4.0 1.7 41
Ficall Paque~ 3.7 1.8 31
CA 02358326 2001-10-O1
-20 -
FULL CITATIONS FOR REFERENCES REFERRED TO IN THE
SPECIFICATION
1. Braun et al., N. Engl. J. Med., 342:525:533.
2. deWynter, E.A. et al., 1975, Stem Cells, Vol. 13:524-532.
3. Firat et al., 1988, Bone Marrow Transplantation, Vol. 21:933-938.
4. Shpall, E.J., et al. 1994, J. of Clinical Oncology 12:28-36.
5. Thomas, T.E., 1994, Cancer Research, Therapy and Control 4(2): 119-
128.
6. Vaughan et al., 1990, Proc. Am. Soc. Clin. Oncol. 9:9.
7. Patel et al., 1995, Clinica Chimica Acta 240: 187-193.
8. Patel and Rickwood, 1995, J. Immunol. Meth. 184: 71-80.
9. Bildirici and Rickwood, 2000, J. Immunol. Meth. 240: 93-99.
10. Bildirici and Rickwood, 2001, J. Immunol. Meth. 252: 57-62.
11. Patel et al., 1993, J. Immunol. Meth., 163:241-251.
12. Van Vlasselaer US Patent 5,648,223
13. Van Vlasselaer US Patent 5,474,687
14. Van Vlasselaer US Patent 5,646,004
15. Van Vlasselaer US Patent 5,840,502
16. Coulter et al. US Patent 5,576,185
17. StemCell Technologies, 1999/2000 Catalogue supplement
