Note: Descriptions are shown in the official language in which they were submitted.
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-1-
SELECTABLE MARKER FOR GENETICALLY ENGINEERED CELLS
AND TISSUES
BACKGROUND OF THE INVENTION
(a) Field of the Invention
The invention relates to novel expression vector for selection of
transduced and/or transfected cells expressing gene of interest. The
invention particularly relates to a selectable marker gene for genetically
engineered cells. Also disclosed is a method of selecting transduced
and/or transfected cells for ex vivo gene therapy.
(b) Description of Prior Art
High gene transfer efficiency and transgene expression in target
primary cells is an essential but challenging aspiration in gene therapy
applications. A remedy may be to dominantly select and enrich ex vivo the
cells concomitantly expressing the therapeutic gene and an attached
selectable marker gene.
Numerous investigators have employed dominant selection
strategies utilising drug resistance genes, such as the prokaryotic neomycin
phosphotransferase II (Neon) drug resistance gene, the human multidrug
resistance (MDR) gene, the mutated human dihydrofolate reductase (DHFR)
gene, and the human 06-methylguanine-DNA-methyltransferase (MGMT)
gene. However, disadvantages have been noted with some drug selectable
markers. For instance, cells modified with the prokaryotic Neon gene may
elicit an immune reaction in vivo, MDR-transduced hematopoietic stem cells,
when transplanted in mice, will lead to a myeloproliferative syndrome and
MGMT genetically engineered cells require selection with the DNA
damaging alkylating agents. Several investigators have revealed the utility of
bone marrow stromal cells, the progenitor cells for nonhematopoietic
tissues, in cell and gene therapy strategies. These autologous cells, which
are easily isolated from bone marrow aspirates, expanded in culture, and
gene modified, have been genetically engineered for the expression of an
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-2-
exogenous gene product and/or for the secretion of therapeutic proteins in
vitro and in vivo.
Numerous investigations have reported the use of drug resistance
genes for in vivo selection of genetically altered cells. The disadvantages of
in vivo drug selection, such as the systemic side-effects of chemotherapeutic
agent administration, are not encountered with in vitro drug selection, since
enrichment of the population of strong transgene expressing cells is
conducted prior to transplantation.
Gene transfer studies regularly employ drug resistance genes for
the in vitro selection of genetically engineered cells. For instance, the
neomycin phosphotransferase 1l (NeoR) drug resistance gene confers to
transduced cells the ability to be selectively expanded through treatment
with neomycin or its analogue 6418. The Neon gene however is not a
suitable selectable marker for clinical applications since it is prokaryotic
and
immunogenic. Moreover, one study reported the adverse efFect of the Neon
gene on the concomitant expression of a second gene product (Apperley et
al., 1991, Blood 78:310-317). °
The human multidrug resistance gene (MDR) which codes for the
multidrug transporter P-glycoprotein has been extensively described and
utilised as a dominant selectable marker with MDR-responsive
chemotherapeutic agents, such as paclitaxel and colchicine (Licht et al.,
2000, Gene Therapy 7:348-358). Nonetheless, some investigators observed
a myeloproliferative syndrome in mice transplanted with MDR-transduced
hematopoietic stem cells.
Another drug resistance gene, the O6-methylguanine-DNA-
methyltransferase (MGMT) gene codes for a DNA repair enzyme that when
overexpressed renders normal mammalian cells resistant to 06-alkylating
agents, such as the nitrosoureas and related methylating compounds. Since
these antineoplastic drugs produce DNA damage which can have toxic,
mutagenic, transforming, and carcinogenic repercussions, their use for
dominant selection of gene-modified cells may be hazardous.
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
_3_
Alternative dominant selectable markers that are well-defined and
widely utilised comprise the mutant variants of the human dihydrofolate
reductase (DHFR) gene which maintain a role in folate metabolism when
introduced into cells but confer drug resistance against the cytotoxicity of
antifolates, such as methotrexate and trimetrexate. Several investigators
have accomplished in vitro drug selection of mutant DHFR-transduced cells.
However, for the untransduced cells to be eradicated by antifolate toxicity,
depletion of thymidine from the cell media is required and may be achieved
through thymidine phosphorylase treatment of the media or by the utilisation
of dialysed thymidine free serum. Moreover, one study demonstrated that
retroviral transfer of the human aldehyde dehydrogenase class-I gene
permits in vitro selection with 4-hydroperoxy cyclophosphamide of
transduced IC562 leukemia cells (Moreb et al., 1998, Human Gene Therapy
9:611-619). Other selectable markers in transfected or transduced cells
include the glutamine synthetase gene, the hygromycin resistance gene, the
zeocin resistance , gene, as well as the puromycin, histidinol D, and
phleomycin drug resistance genes.
Retroviral transfer of a selectable drug resistance gene may in
some cases confer lower expression of the therapeutic transgene or raise an
immune reaction against gene-modified cells, ultimately reducing the
effectiveness of the gene therapy approach. The immunogenicity of certain
drug selectable genes such as Neon, led to their removal.
It would be highly desirable to be provided with an efficient
selectable marker gene which allow for preparing transduced or non
immunogen transfected cells producing a protein of interest, and for ex
viva gene therapy and protein delivery.
SUMMARY OF THE INVENTION
An object of the present invention is to provide an expression
vector for identifying transduced and/or transfected cells.
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-4-
In accordance with the present invention, there is provided an
expression vector for identifying transduced and/or transfected cells
expressing a nucleotitic sequence of interest. The expression vector
comprises:
- a suitable promoter compatible with the transfected cells;
- an internal ribosome entry site (IRES);
- a marker nucleotidic sequence encoding drug resistance
cytidine deaminase (CD), a functional fragment or an analog
thereof, operably linked to the IRES; and
- an endogenous or exogenous nucleotidic sequence of
interest operably linked to the IRES;
wherein the marker nucleotidic sequence when detected in cells, indicates
that the cells are transduced and/or transfected and capable of expressing
the nucleotidic sequence of interest, the nucleotidic sequence encoding
drug resistance cytidine deaminase (CD), a functional fragment or an
analog thereof, being operably linked upstream of the IRES when the
nucleotidic sequence of interest is operably linked downstream of the
IRES, and the nucleotidic sequence encoding drug resistance cytidine
deaminase (CD), a functional fragment or an analog thereof, being
operably linked downstream of the IRES when the nucleotidic sequence of
interest is operably linked upstream of the IRES.
The expression vector of the invention may further be flanked by
retroviral long terminal repeat (LTR) sequence at 5' and/or 3' ends of the
vector.
The expression vector may be composed of DNA or RNA
selected from the group consisting of eukaryotic, viral, adenoviral, adeno-
associated, Simliclei, and Herpes simplex expression vectors.
The cells of the present invention may be selected from the
group consisting of stromal, epithelial, fibroblasts, myoblasts, muscular,
stem, progenitor, blood, and hematopoietic cells.
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-5-
The nucleotidic sequence of interest may for example encodes a
protein selected from the group consisting of cytokine, interleukin, growth
hormones, hormones, blood factors, marker proteins, immunoglobulins,
antigens, releasing hormone, anticancer product, antiviral product,
antiretroviral product, an antisense, an antiangiogenic product, and a
replication inhibitor.
The promoter may be a cytomegalo-virus (CMV) promoter.
In accordance with the present invention there is also provided
an expression vector, wherein the nucleotidic sequence encoding for drug
resistance cytidine deariiinase (CD), a functional fragment or an analog
thereof.
In accordance with the present invention there is provided a host
cell transduced and/or transfected with the expression vector of the
invention, and uses thereof in the preparation of a medicament for ex vivo
treatment.
Also in accordance with the present invention, there is provided
a method of identifying genetically transduced or transformed cells
expressing a nucleotidic sequence of interest. The method comprises the
steps of:
a) providing an expression vector for identifying transduced
and/or transfected cells expressing a nucleotidic sequence of interest, the
expression vector comprising:
- a suitable promoter compatible with the transfected cells;
- an internal ribosome entry site (IRES);
- a marker nucleotidic sequence encoding for drug resistance
cytidine deaminase (CD), a functional fragment or an analog
thereof, operably linked to the IRES; and
- a nucleotidic sequence of interest operably linked to the
IRES;
wherein the nucleotidic sequence encoding drug resistance cytidine
deaminase (CD), a functional fragment or an analog thereof, being
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-6-
operably linked upstream of the IRES when the nucleotidic sequence of
interest is operably linked downstream of the IRES, and the nucleotidic
sequence encoding drug resistance cytidine deaminase (CD), a functional
fragment or an analog thereof, being operably linked downstream of the
IRES when the nucleotidic sequence of interest is operably linked
upstream of the IRES;
b) transducing and/or transfecting cells with the expression
vector of step a);
c) culturing the cells of step b) under conditions suitable for the
expression vector to express the marker nucleotidic sequence
encoding the drug resistance cytidine deaminase, and the
nucleotidic sequence of interest, the nucleotidic sequence of
interest coding for a protein of interest; and
d) treating cells with cytosine nucleoside analogs;
wherein the cells living after the treating step d) are indicative that the
cells
are transduced and/or transfected and capable of expressing the
nucleotidic sequence of interest.
For the purpose of the present invention the following terms are
defined below.
The term "functional fragment " as used herein is intended to
mean any portion of the gene encoding for a peptide or protein having the
activity of the cytidine deaminase (CD).
The term "analogs " as used herein is intended to mean any
functional modified form of the cytidine deaminase. The modification may
become from a mutated or modified form of the nucleotidic sequence
encoding CD.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 is a schematic illustration of the murine stem cell virus
(MSCV) derived retroviral plasmids;
Fig. 2 is a schematic illustration of DNA integrated retrovector;
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-7-
Fig. 3 illustrates southern blot analysis of CD-IRES-EGFP
transduced A549 cells;
Fig. 4 illustrates the effect of ARA-C on the proliferation of
genetically engineered A549 cells;
Fig. 5 illustrates the efFect of dFdC on the proliferation of
genetically engineered A549 cells;
Fig. 6 illustrates the resistance of CD gene-modified primary
human lymphocytes to growth inhibition by ARA-C;
Fig. 7 illustrates the percent GFP positive cells assessed by flow
cytometry analysis after in vitro enrichment of CD-IRES-EGFP transduced
primary mouse marrow stromal cells;
Fig. 8 illustrates the flow cytometry analysis of one representative
experiment demonstrating dose-dependent selective expansion of CD-
IRES-EGFP transduced stromal cells; and
Fig. 9 illustrates the mean GFP fluorescence plotted against ARA-
C concentration. Average ~ SEM (n=3).
DETAILED DESCRIPTION OF THE INVENTION
In accordance with the present invention, there is provided a
drug resistance gene (CD) that can serve as an in vitro or ex vivo positive
selectable marker in genetically engineered cells, preferably primary
autologous cells.
In one embodiment of the present invention, there is provided a
human cytidine deaminase gene for ex vivo selective enrichment of
genetically engineered primary cells, for cell and gene therapy purposes.
The present invention also provide with an efficient selectable marker when
linked to therapeutic exogenous nucleic acid sequences for treatment of
genetic, malignant, immune, and viral diseases. For example, for patients
implanted with ex vivo drug selected cells. the method of the present
invention would provide engrafted cells constituted entirely by genetically
engineered cells, thus improving therapeutic outcome.
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-g_
In another embodiment of the invention, there is provided a drug
resistance gene with potential as a novel drug selectable marker. This drug
resistance gene is the human cytidine deaminase (CD) gene. CD catalyses
the deamination of cytidine or deoxycytidine to uridine or deoxyuridine,
respectively and in addition, can deaminate cytosine nucleoside analogues
causing their pharmacological inactivation. The CD drug resistance gene
has recently shown promise for protecting normal hematopoietic cells from
the dose-limiting myelotoxicity of anti-cancer drugs to thereby permit dose
escalation for enhanced chemotherapeutic effectiveness. This
chemoprotection approach using the human CD gene is the antithesis of a
suicide gene therapy strategy utilising the prokaryotic cytosine deaminase.
Unlike human CD or other optimal drug resistance genes, suicide genes
code for non-mammalian enzymes, such as Escherichia coli cytosine
deaminase and Herpes Simplex Virus Thymidine Kinase (HSV-TK), which
are introduced into malignant cells to convert inactive prodrugs, such as 5-
fluorocytosine and gancyclovir, respectively, into cytotoxic agents.
In the present invention, the human CD cDNA which consists of
910bp encoding a 146 amino acid protein of 48.7 kilodaltons was cloned and
expressed. The small size of the CD cDNA is a favourable feature endorsing
its utility for gene therapy applications. It has demonstrated enhanced CD
expression as well as cytosine nucleoside analogue drug resistance
pursuant to retroviral gene transfer of the CD cDNA in mouse fibroblast and
primary hematopoietic cells. Furthermore, CD proviral DNA persistence and
long-term expression in various tissues has been determined in mice
transplanted with CD gene-modified hematopoietic cells.
The human CD gene of the present invention possesses assets
as a drug selectable marker. It is not a mutated variant nor prokaryotic,
thereby indicating that drug selected CD gene-modified cells when
implanted in vivo will not induce an immune response. Moreover, it has a
small coding sequence and the agents that it efficiently confers resistance
to, and which are to be utilised for ex vivv selection, are antimetabolites
that
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
_g_
will not cause DNA damage in CD-enriched cells. The main mechanism of
action of ARA-C, an effective antileukemic drug, and dFdC, a promising
antitumor agent, is via inhibition of DNA synthesis, whereas that of 5-AZA-
CdR, which has shown antileukemic and interesting antitumor activity, is via
inhibition of DNA methylation and ensuing activation of tumor suppressor
genes. The human CD gene may serve as an ex vivo positive selectable
marker with cytosine nucleoside analogues in genetically engineered
primary autologous cells,
There is also provided with the invention an efficient in vitro drug
selection of gene-modified primary marine marrow stromal cells. Bone
marrow constitutes a source of stem cells for hematopoietic progenitors, and
a provenance of mesenchymal stem cells or marrow stromal cells which
have the ability to differentiate into several cell types including
chondrocytes,
osteoblasts, adipocytes, myoblasts, cardiomyocytes, and astrocytes. Marrow
stromal is thus an autologous tissue that .is an ideal target for cell and
gene
therapy strategies due to its capacity for self-renewal and differentiation.
Dominant selection of gene-modified marrow stromal cells ex vivo
may allow a greater proportion of cells to express therapeutically relevant
levels of the beneficial gene product in vivo thus enhancing clinical
effectiveness.
In one embodiment af,the present invention, there is provided
bicistronic vector containing CD or the GFP reporter gene as marker
operably linked to a gene encoding a protein of interest, generated
retroparticles, and human A549 cells as well as primary human lymphocytes
that can be transduced with these retrovirions in vitro and consequently may
acquire the cytosine nucleoside analogue drug resistance or green
fluorescence phenotype as indicators of efficient expression of the protein of
interest. Primary marine marrow stromal cells may be also transduced and
showed in vitro dose-dependent selection of CD gene-modified stromal cells
using ARA-C.
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-10-
Therefore, the introduction of a dominant selectable marker in
marrow stromal allows selection and enrichment of genetically engineered
cells ex vivo to ensure sufficient levels of the therapeutic protein for a
favourable clinical outcome. The human CD gene possesses the potential to
enhance the proportion of gene modified marrow stromal cells expressing a
therapeutic transgene, such as type I collagen in children with osteogenesis
imperfect. Furthermore, CD gene transfer may permit the ex vivo expansion
of cells genetically engineered to express optimal amounts of therapeutic
soluble proteins such as Factor VIII and Factor IX in hemophilia and
Erythropoietin in hemoglobinopathies, such as ~3-thalassemia. Marrow
stromal cells may thus serve as a gene delivery vehicle in diseases where
clinical improvement is possible through the systemic secretion of a
therapeutic gene product.
In one embodiment, target primary cells for selectable marker
gene transfer may include bone marrow stromal cells and lymphocytes but
also hematopoietic cells, myoblasts and/or fibroblasts. Gene transfer of
human CD into hematopoietic stem cells and/or lymphocytes may be
valuable in the treatment of disorders that afflict the hematopoietic system,
such as adenosine deaminase deficiency and chronic granulomatous
disease, as well as storage disorders such as Gaucher disease and Hunter
syndrome. Additionally, CD for ex vivo selection may be used to augment
the proportion of hematopoietic cells transduced with anti-HIV-I genes, such
as RevMlO, and consequently give rise to an enriched population of mature
T-lymphocytes and monocytic cells with high level antiviral gene expression.
Furthermore, the human CD gene may serve as a dominant
selectable marker in cancer gene therapy applications employing a non-
selectable therapeutic transgene such as a tumor-suppressor gene to inhibit
tumor growth or a cytokine gene to strengthen the immune response against
neoplastic cells. The human CD gene may also be utilised as a positive
selectable marker for the therapy of autoimmune diseases such as arthritis,
systemic lupus erythematosus, and colitis, by selectively enhancing cells
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-11 -
genetically engineered to express regulatory cytokine genes such as IL-4
and IL-10, or inflammatory cytokine inhibitory genes such as 1L-I, IL-2, TNFa,
and IFNy.
In contrast to selectable markers of prokaryotic origin, the CD
gene, because of its human origin is not anticipated to be immunogenic in
clinical applications, and as it is demonstrated in the present invention, has
not impeded but enhanced the expression of a second transgene from a
bicistronic retrovector.
MATERIALS and METHODS
Cell Lines
GP+E86 ecotropic and GP+envAM12 amphotropic retrovirus
packaging cell lines, kindly supplied by A. Bank (Columbia University, New
York) were maintained in Dulbecco's modified essential medium (DMEM)
(Canadian Life Technologies, Burlington, Ontario) supplemented with
10°!°
heat-inactivated foetal bovine serum (FBS) (Wisent Technologies, St.
Bruno, Quebec) and 5~,glml gentamicin (Canadian Life Technologies)
The 293GPG retrovirus-packaging cell line, generously provided
by R.C. Mulligan (Children's Hospital, Boston, MA) was grown in 293GPG
media [DMEM supplemented with 10% heat-inactivated FBS, 0.3mg/ml
6418 (Mediatech, Herndon, VA), 2~,g/ml puromycin (Sigma, Oakville,
Ontario), 1 ~.g/ml tetracycline (Fisher Scientific, Nepean, Ontario), and
50units/ml penicillin/streptomycin (Pen/Strep) (Wisent)].
A549 human lung carcinoma cells, obtained from American
Type Culture Collection (ATCC), were maintained in RPMI 1640 medium
(Canadian Life Technologies) supplemented with 10% heat-inactivated
FBS.
NIH 3T3 mouse fibroblast cells, from ATCC, were cultured in
DMEM with 10% FBS and 50U/ml Pen/Strep. All cells were grown in a
humidified incubator with 5% C02. (All cells were incubated at 37°C
with
5% C02).
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
_12_
Retroviral vector design and virus-producing cell line generation
The murine stem cell virus (MSCV) derived retroviral vector
pIRES-EGFP, a bicistronic construct comprising a multiple cloning site linked
by an internal ribosomal entry site (IRES), to the enhanced green
fluorescent protein (EGFP) reporter, was kindly provided by Robert Hawley.
The retroviral vector pCD-IRES-EGFP (Fig. 1, pIRES-EGFP comprises a
multiple cloning site (mcs) and the EGFP reporter gene parted by an internal
ribosomal entry site (fRES). pNeoR-IRES-EGFP is a derivative of pIRES-
EGFP where the Neon gene is inserted in the mcs. Likewise, pCD-IRES-
EGFP contains the human CD cDNA in. the mcs upstream of the IRES) was
synthesised by retrieving the human cytidine deaminase (CD) cDNA
sequence by Ncol/Klenow and BamHl digest of the pMFG-CD construct,
and ligating it with a Xhol/Klenow and BamHl digest of pIRES-EGFP. The
retrovector pNeoR-IRES-EGFP, encompassing the neomycin
phosphotransferase II drug resistance gene, was similarly constructed to
serve as an additional control plasmid.
The pCD-IRES-EGFP vector (10pg) was introduced into GP+E86
packaging cells by calcium phosphate transfection utilising the Cell Phect~'
kit (Pharmacia) and cells were subsequently selected for 3 weeks in
complete media supplemented with 2.5p,M cytosine arabinoside (ARA-C)
(Upjohn, Don Mills, Ontario). The ensuing stable polyclonal producer cell
population was used to supply virus for the transinfection (or transduction)
of
GP+envAM12 amphotropic packaging cells. Briefly, for 3 consecutive days,
retrovirus-containing supernatant was harvested from subconfluent ecotropic
producer cells, filtered with a 0.45pm syringe mounted filter (Gelman
Sciences, Ann Arbour, MI) and applied with 8pg/ml Polybrene~' (Sigma
Chemical, St. Louis, MO) over target GP+envAM12 cells. Three days later,
these amphotropic producers commenced a 2-week drug selection by
culturing in media including 2.5pM ARA-C, thence generating the polyclonal
population GP+envAM12-CD-IRES-EGFP.
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-13-
As a control, the GP+AM12-Neon-IRES-EGFP polyclonal
producer was created in an almost identical manner to that described above,
the exception being that drug selection was performed with 400~,g/ml G418-
containing media. In addition, GP+E86-Lac Z cells were generated by
transinfection of the GP+E86 cell line with filtered retroviral supernatant
from
the 293GPG-LacZ producer (kind gift from R.C. Mulligan, Children's
Hospital, MA) twice a day for 3 consecutive days. Moreover, utilising
Lipofectamine~' (Gibco-BRL, Gaithesbug, MD), the pantropic 293GPG
packaging ,cell line is co-transfected with 5~.g pCD-IRES-EGFP and 70ng
pJ6S2Bleo graciously given by R.C. Mulligan (Children's Hospital, MA).
These cells then underwent 4-week selection in 293GPG media
supplemented with 100~,g/ml ZeocinT"~ (Invitrogen, San Diego, CA)
consequently generating the stable polyclonal producer 293GPG-CD-IRES-
EGFP. By this same approach, the control 293GPG-IRES-EGFP producer
was also conceived. Flow cytometry analysis of cells for GFP expression
was conducted using an Epics XL/MCL Coulter analyzer and gating live cells
based on FSC/SSC profile. Retroparticles from virus producers were noted
to be free from replication competent retrovirus (RCR) by GFP marker
rescue assay employing supernatant from transduced target cells.
Titration of retrovirus-producers
To determine the titer of GP+E86 and GP+envAM12 producers,
NIH 3T3 cells were utilised, whereas A549 cells were used for titering
293GPG producers. These target cells were plated at a density of 2 to 4 x
104 cells per well in 6-well tissue culture dishes and the following day,
cells
from one test well were trypsinized and counted to ascertain the baseline
cell number at time of virus addition. For transduction, serial dilutions of
retroviral supernatants (0.01 to 1001 in a final volume of 1 ml complete
media, supplemented with 8gg/ml polybrene, were placed over the adherent
target cells. Flow cytometry analysis was realised 72 hours post-transduction
to disclose the percentage of GFP-expressing cells. The titer was calculated
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-14-
utilising the equation below by considering the virus dilution that led to 10-
40% GFP positive cells.
Titer (infectious particles/ml) _ (% GFP positive cells) x (number of
cells at initial virus exposure)/(volume of virus in the 1 ml applied to
cells).
Transduction of human A549 cells analysis and drug selection
Retroviral supernatant from 293GPG-CD-IRES-EGFP cells grown
to confluence in 293GPG media devoid of tetracycline for over 72 hours to
allow VSVG-pseudotyped retroparticle production were placed in a 25cm2
flask of subconfluent A549 cells with 3~,g/ml PolybreneT"~. This transduction
procedure was executed once a day for 3 consecutive days yielding A549-
CD-IRES-EGFP cells. As a control, A549 cells were likewise transduced with
viral particles from 293GPG-IRES-EGFP producers hence giving rise to
A549-IRES-EGFP cells. Five days following the last transduction round, flow
cytometry analysis was performed to determine gene transfer efficiency and
transgene expression as evaluated by GFP fluorescence. Stably transduced
A549-CD-IRES-EGFP cells were subsequently expanded in complete media
only, as well as in the presence of two different concentrations of ARA-C,
1 ~,M and 2.5~,M, for 12 days.
Southern Blot Analysis
Genomic DNA was isolated from CD-IRES-EGFP stably
transduced A549 cells untreated or treated with ARA-C as well as from
control cells, using a QIAGEN'" genomic DNA isolation kit. For Southern blot
analysis, 5~,g of genomic DNA was digested with Nhel and separated by
~ agarose gel electrophoresis. Following UV photography with a fluorescent
ruler, the gel was immersed in denaturing solution (0.5M NaOH; 1.5M NaCI)
and then in neutralising buffer (0.5M Tris-HCI pH7; 1.5M NaCI, pH7), each
for 45 minutes. The DNA in the gel was transferred onto a Hybond-NTM
nylon membrane (Amersham, Oakville, Ontario) using 10X Standard Saline
Citrate (SSC) for an approximately 48 hour downward transfer with the
Turbo Blotter'" device (Schleicher & Shuell, Keene, NH). The membrane
was then irradiated in a Bioslink~' UV linker with 0.3 J/cm2 and hybridised in
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-15-
Express Hyb~' solution (Clontech) containing PCR-amplified 32P-labeled
cDNA for human CD. The membrane was washed and exposed using a
Phosphor Imagery'.
Growth inhibition assay (MTT) of Gene-modified A549 cells
, Gene-modified A549 cells were plated in 96-well flat-bottom tissue
culture dishes at a density of 1500 cells per well. Various concentrations of
ARA-C or dFdC (Lilly Research Laboratories, Indianapolis, Indiana) were
added to cells in a final volume of 1001 RPMI/10% FBS. Cells were placed
at 37°C and 4 days later exposed to 20,1 of a solution containing 3-
(4,5-
deimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, i.e. MTT, for 4 hours
at 37°C, and subsequently to 100p,1 of a solubilization buffer (50% N,N-
dimethylformamide, 20% SDS) for approximately 15 hours. The optical
density was then measured at 570nm with reference at 630nm using a multi-
well plate reader, thus providing an assessment of the percentage of
surviving cells based on the cleavage of MTT by the mitochondrial
dehydrogenases of living cells to the detectable formazan. Percent cell
survival was ascertained using the following equation:
~OD570I630 test well - OD57o/630 media Only well] x 100.
cell survival =
[OD5~oi63o cells only well - OD5~o~63o media only well]
CD enzyme assay
Stably transduced test and control A549 cells were assayed for
CD enzyme activity. Concisely, these adherent cells (2-5 x 10' cells) were
trypsinized, washed with phosphate-buffered saline (PBS) and resuspended
in 5mM Tris-HCI (pH 7.4) and 5mM dithiothreitol. The cell suspension was
freeze-thawed rapidly three times, centrifuged at high speed (12 000 rpm)
for 15 minutes and the supernatant consisting of the cytosolic extract was
collected. For CD enzyme assay, various dilutions of the cytosol were
utilised in a reaction mixture with 50mM Tris-HCI and 0.5~.Ci 3H-cytidine (ICN
Biomedicals, Irvine, California). The reaction was allowed to proceed for 30
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
- 16-
minutes at 37°C and then stopped with cold HCI (0.001 N). The mixture
was
subsequently poured onto Whatman~' P-81 phosphocellulose discs and the
quantity of radioactivity bound to the 'discs determined by scintillation
counting. Oniy cytosine nucleosides, and not the deaminated uracil
nucleosides, are bound to the discs. One unit of enzyme activity was defined
as the amount of enzyme that catalyses the deamination of one nmole of
3H-cytidine per minute at 37°C. Total protein concentration was
determined
using the Bio-Rad~' protein assay (Bio-Rad Laboratories, Mississauga,
Ontario) with bovine serum albumin as the standard.
Transduction of primary human lymphocytes and Growth suppression assay
Primary human lymphocytes were isolated from peripheral blood
by Ficoll Hypaque~' centrifugation and plated at a cell density of ~1 x 106
cells/ml in RPMI supplemented with 10% FBS, 50U/ml Pen/Strep, and
200U/ml human IL-2. The next day, lymphocytes were activated with
phytohemagglutinin (PHA) for 72hrs and subsequently co-cultivated over.
subconfluent 293GPG-CD-IRES-EGFP or control 293GPG-IRES-EGFP
virus producers tetracycline withdrawn three days earlier for optimal high
titer
virion generation. Lipofectamine~' was added for a final concentration of
6~g/ml, and co-cultivation permitted to proceed for 48 hours. Lymphocytes
were then carefully collected and three days later plated at a density of 600
000 cells per well of 48-well tissue culture dishes in the absence or presence
of various concentrations of ARA-C in a final volume of 800p,1 complete
media. After a two week period, viable cells were enumerated by trypan blue
exclusion and cell survival measured as such: % cell survival = (number of
cells in test well/number of cells grown in media only) x 100. (Flow cytometry
analysis for CD3 expression was conducted and all cells confirmed to be
lymphocytes.)
Harvest expansion and transduction of primary/ mouse bone marrow stoma
One female C57B1/6 mouse was sacrificed by C02 inhalation and
bone marrow cells harvested by flushing the hind leg femurs and tibias with
DMEM~' supplemented with 10% FBS and 50Ulml Pen/Strep. Whole
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-17-
marrow was plated in 10cm (diameter) tissue culture dishes and five days
later, the nonadherent hematopoietic cells were discarded and the adherent
bone marrow stromal cells cultured (for ~3 months) at 37°C with 5% C02.
Supernatant containing virions from subconfluent GP+envAM12-CD-IRES-
EGFP or from control GP+envAM12-Neon-IRES-EGFP producers was
placed over marrow stromal cells plated at subconfluence. Transduction was
carried out twice a day for three successive days in the presence of 6pg/ml
LipofectamineTM. Over 72 hours pursuing to the last transduction round, flow
cytometry analysis for GFP expression was performed to evaluate the gene
transfer efficiency.
In vitro selection and enrichment assay of CD Gene-modified marrow stoma
CD-IRES-EGFP transduced stromal cells were mixed in various
proportions with untransduced stromal cells and plated at a density of 25
000 cells/well of 6-well plates without ARA-C and with ARA-C at a final
concentration of 2.5wM. Drug exposure was ceased 7 days later, and cells
expanded for an additional week in DMEMn'' with 10°!° FBS and
50U/ml
Pen/Strep and analysed by flow cytometry analysis. Moreover, the cell
combination where CD-IRES-EGFP modified stroma constituted 10% of
total cell number was plated at 25 000 cells/well in 6-well dishes with a
range
of ARA-C concentrations. Pursuant to a 1-week drug exposure (preventing
total confluence by cell passaging) and subsequent week of cell expansion,
percent GFP positive cells and mean GFP fluorescence were evaluated by
flow cytometry analysis.
RESULTS
GFP expression and titer determination of virus-producers
In order to evaluate gene transfer efficiency and transgene
expression in~stably transfected and transinfected virus producers, flow
cytometry analysis for GFP expression was conducted. The percentage of
GFP positive cells in the polyclonal producer populations GP+envAM12-CD-
IRES-EGFP, GP+envAM12-Neon-IRES-EGFP, 293GPG-CD-IRES-EGFP,
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-18-
and 293GPG-IRES-EGFP were revealed to be 99.2, 100, 50.4, and 65.2,
respectively, as compared to values of less than 3% for parental unaltered
cells. To estimate the amount of infectious particles released by these cells,
their retroviral supernatant was utilised in a titration assay and the
extrapolated viral titers were 1.5 x 106, 1.3 x 106, 1 x 106, and 0.5 x 106
infectious particles per ml. (The titer of LacZ-GP+E86 virus producers
estimated through X-gal staining of transduced NIH 3T3 cells, was ~1.1 x
105 infectious particles/ml.)
GFP e~ression of CD gene-modified A549 cells
Flow cytometry analysis was first executed on A549 cells
succeeding their transduction and the percentage of cells expressing GFP
was 95.9 for A549-CD-IRES-EGFP cells and 92.4 for A549-IRES-EGFP
cells, in contrast to 0.2 for untransduced A549 cells.
Southern blot analysis
In order to demonstrate that the recombinant retroviral construct
CD-IRES-EGFP did not sustain any rearrangements or deletions before
integration as proviral DNA into the genome of transduced cells, Southern
blot analysis was carried out on gene-modified A549 cells. This procedure
revealed a DNA band corresponding to the 3683bp fragment expected from
Nhel digest of integrated unrearranged CD-IRES-EGFP proviral DNA (Fig.
3). Genomic DNA from the indicated cell lines was digested with Nhel, which
cuts once in each flanking LTR, and fractionated on 1 % .agarose gel.
Hybridisation of the blot with a 32P-labeled CD cDNA probe permits detection
of integrated, unrearranged proviral DNA of the predicted 3683bp size.
Molecular weights are indicated). In A549-CD-IRES-EGFP cells selected
with 2.5~,M ARA-C, the DNA band detected was of higher intensity than that
of cells exposed to 1 ~,M ARA-C, and notably more so than that of CD-
transduced cells which were not drug selected (Fig. 3).
Expression of CD enzyme activity in gene-modified A549 cells
To evaluate the CD activity in genetically engineered cells,
enzyme assay was performed on the cytosolic extract of A549 cells
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-19-
transduced with CD-IRES-EGFP retroviral particles and on that of control
cells, i.e. untransduced A549 cells, as well as A549 cells modified with
IRES-EGFP virions. As detailed in Table 1, parental A549 cells as well as
A549-IRES-EGFP cells showed low CD activity, specifically 0.29 ~ 0.07 and
0.53 ~ 0.07 Units/mg of protein, respectively. In contrast, the CD activity of
A549-CD-IRES-EGFP cells was 515 ~ 31 Units/mg of protein, representing
an over 1000=fold augmentation. CD enzyme assay was also conducted on
the cytosol obtained from CD-IRES-EGFP transduced A549 cells exposed
to 2.5p,M ARA-C for 12 days, and the CD enzyme activity measured was
524 ~ 41 Units/mg of protein. In Table 1, CD activity was measured in
cytosolic extracts of the different cell lines. Units of enzyme activity are
defined as nmoles of cytidine deaminated per minute.
Table 1
Cytidine deaminase (CD) activity in gene-modified A549 cells.
Cell line Enzyme Activity
A549 0.29 ~ 0.07
A549-IRES-EGFP 0.53 ~ 0.07
A549-CD-IRES-EGFP 515 ~ 31
A549-CD-IRES-EGFP 524 ~ 41
(selected in 2,5 mM ARA-C)
Values represent average ~ SEM, n= 4-12.
Growth inhibition by ARA-C and dFdC of Gene-modified A549 Cells (or:
Drua sensitivity of gene-modified A549 cells)
The sensitivity of genetically engineered A549 cells to the toxicity
of cytosine nucleoside analogues ARA-C and dFdC was evaluated by MTT
assay. As depicted in Fig. 8 (Parental A549 cells, A549 cells transduced with
control IRES-EGFP retroparticles, and A549 cells transduced with CD-IRES-
EGFP virions and expanded without ARA-C, with 1 ~M ARA-C, or 2.5~,M
ARA-C for 12 days, were subsequently exposed to ARA-C for 4 days and
cell survival quantified by MTT assay. Percent survival is plotted against
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-20-
drug concentration (log scale). Average ~ SEM, n >_ 8), ARA-C at a
concentration of 1 p,M caused a substantial decrease in the % cell survival of
untransduced A549 and A549-IRES-EGFP cells to values of 15.3 ~ 0.6 and
10.6 ~ 0.4, respectively. However, at this same drug concentration, survival
by A549-CD-IRES-EGFP cells was over 3-fold greater at 47.4 ~ 1.9%. To
demonstrate a relationship between in vitro drug selection pressure and
resulting drug resistance magnitude, the sensitivity of the A549-CD-IRES-
EGFP cells previously treated with 1~,M, as well as 2.5p,M ARA-C was also
assessed. At the ARA-C concentration of 1 ~,M, the % survival of the 2.5p,M
ARA-C-selected CD-modified cells was 70.7 ~ 1.3, thus significantly higher
than the above-mentioned 47.4 ~ 1.9% cell survival of A549-CD-IRES-EGFP
cells which had not undergone post-transduction drug selection (Fig. 4).
To ascertain that these CD gene-modified human cells are not
resistant uniquely to ARA-C but have conjointly acquired cross-resistance to
other cytosine nucleoside analogues, growth suppression by dFdC was also
evaluated. Treatment with 0.1 wM dFdC greatly suppressed cell survival of
control A549 and A549-IRES-EGFP cells to 15.1 ~ 0.5% and 10.7 ~ 0.4%,
respectively (Parental A549 cells, A549 cells transduced with control IRES-
EGFP retroparticles, and A549 cells transduced with CD-IRES-EGFP virions
and expanded without ARA-C, with 1 ~,M ARA-C, or 2.5p,M ARA-C for 12
days, were subsequently exposed to dFdC for 4 days and cell survival
quantified by MTT assays. Percent survival is plotted against drug
concentration (log scale). Average ~ SEM, n >_ 8). In contrast however,
A549-CD-IRES-EGFP cells demonstrated considerably less drug sensitivity,
with 86.4 ~ 2.4% survival at the same concentration of dFdC. Likewise, the
2.5pM ARA-C selected CD-IRES-EGFP modified cells revealed at 0.1 pM
dFdC, 92.1 ~ 1.7% cell survival. Furthermore, treatment with the highest
concentration of 1 pM dFdC considerably reduced cell survival of unselected
A549-CD-IRES-EGFP cells to 17.2 ~ 0.7% whereas the 2.5~,M ARA-C
selected cells showed 53.3 ~ 1.8% survival (Fig. 5).
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-21 -
Effect of ARA-C on growth of genetically-engineered primary human
Iymphocytes
In vitro growth inhibition assay was performed to evaluate the
sensitivity of gene-modified primary human lymphocytes to increasing
concentrations of ARA-C. As illustrated in Fig. 6, treatment with 1 wM ARA-C
practically abolished cell survival of human lymphocytes transduced with
IRES-EGFP retrovirions (5.5 ~ 1.7% survival), whereas did not significantly
affect that of lymphocytes transduced with CD-IRES-EGFP viral particles
(87.5 ~ 7.2% cell survival). In addition, at the ARA-C concentration of 5~.M,
growth of IRES-EGFP-modified lymphocytes was suppressed (0.1 ~ 0.1%),
in vast contrast to CD-IRES-EGFP transduced lymphocytes which
demonstrated virtually complete drug resistance (98.1 ~ 4.2% survival).
Ex vivo enrichment of CD gene-modified primary mouse marrow stromal
cells
Flow cytometry analysis was initially conducted on primary mouse
bone marrow stromal cells 72 hours following their transduction, and the
percentage of GFP positive cells was 98.4% for stroma engineered with CD-
IRES-EGFP retroparticles and 96.3% for cells modified with Neon-IRES-
EGFP virus, as compared to 2.1 % for untransduced stromal cells.
To determine if in vitro selection and enrichment of CD-expressing
stroma can occur, CD-IRES-EGFP transduced stromal cells were mixed at
10, 20, 30, and 100% proportions with untransduced stroma, cultured in
2.5p.M ARA-C for 7 days, and expanded for one additional week.
Subsequently, flow cytometry analysis revealed, as indicated in Fig. 7,
selective expansion of GFP expressing cells to over 99% for all cell
combinations exposed to ARA-C. (This ARA-C concentration of 2.5~,M was
noted to completely eradicate untransduced as well as Neon-IRES-EGFP
modified stroma). The average green fluorescence of CD-IRES-EGFP
engineered stromal cells also rose ensuing drug selection (Fig. 7).
Moreover, to determine if in vitro enrichment of CD gene-modified
marrow stroma is dose-dependent, a cell mixture consisting of 10% CD-
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-22-
IRES-EGFP vector-bearing stromal cells within a population of untransduced
marrow stroma was exposed to increasing concentrations of ARA-C for one
week and subsequently cultured for another week. Flow cytometry analysis
revealed that the degree of selective expansion of the GFP positive cells
~ was dependent on the dose of ARA-C (Fig. 6). As illustrated in Fig. 8, the
11.7% GFP + cells without ARA-C escalated to 51.5% and 99.6% when .
selected in 0.25 and 2.5~,M ARA-C, respectively (Fig. 8). Drug dose had an
impact on the level of transgene expression, as perceived by the mean GFP
fluorescence rising from 13.0 ~ 3.0 in the absence of ARA-C to 48.3 ~ 6.5
and 99.3 ~ 3.3 in cells selected in 1 and 2.5~,M ARA-C, respectively (Fig. 9).
In vivo enaraftment of ex vivo drug selected Gene-modified stroma
To determine that ex-vivo drug selection of CD engineered stroma
does not alter in vivo engraftment capacity, 10% CD-IRES-EGFP positive
mouse stroma enriched to >99% by ARA-C exposure was thereafter gene-
modified to also express a-galactosidase and implanted by intramuscular
injection in immunocompetent syngeneic mice. As evidenced in the present
experiment, sections of muscle harvested 2 weeks post-implantation contain
engrafted ex vivo drug selected, transgene expressing cells.
Retroviral gene transfer permits stable and efficient integration of
a foreign gene in the chromosomal .DNA of the host cell and subsequently in
all progeny cells. However, high level and long term expression of a
beneficial exogenous gene is difficult to achieve in gene therapy studies.
Nevertheless, gene therapy of many hereditary and acquired affliction
commands that the desired therapeutic gene be efficiently translated in the
majority of target cells. Accordingly, it is anticipated that if the
genetically
engineered cells represent only a small portion of the target cells, their
outnumbering by the unaltered cells will ultimately ensue. One possible
means to overcome this obstacle may be to expand the proportion of gene-
modified cells in vitro through dominant selection relying on the co-
expression of a drug resistance gene.
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-23-
A bicistronic retroviral vector enclosing the human CD cDNA and
the GFP reporter gene is generated. It is noted that its expression lead to
functional levels of CD in genetically engineered human cell line A549,
primary human lymphocytes, and primary murine bone marrow stromal
cells. The convenience of this CD-IRES-EGFP retroviral construct (Fig. 1 ),
where GFP represents the non-selectable therapeutic gene, was the
accorded ability to track the gene-modified cells via the green fluorescence
(emitted at 507nm following blue light excitation at 488nm) which was
visualised by fluorescence microscopy and quantified by flow cytometry
analysis. Therefore, GFP expression here reflected CD expression.
The CD-IRES-EGFP retrovector is integrated as an intact
proviral DNA in the genome of transduced cells, since no rearrangements,
nor deletions are revealed by Southern blot analysis of gene-modified cells
(Fig. 3). The more intense cDNA signal detected with CD gene-modified
A549 cells selected with 2.5~,M ARA-C, versus that with 1 ~,M ARA-C
treated cells, or the weakest band seen with unselected cells, indicates an
augmented copy number of the proviral sequences. This higher copy
number suggests selection and enrichment of the CD positive cells with
multiple integrants and/or amplification of the CD gene by ARA-C exposure
of genetically-engineered cells. It was discovered that it is possible to
increase CD expression via amplification of the proviral CD gene with
ARA-C exposure of CD-transduced fibroblast cells.
Quantification of functional CD expression in CD-IRES-EGFP
transduced A549 cells, as compared to corresponding control IRES-EGFP-
modified cells, disclosed an over 1000-fold increment (Table 1 ). Thus, CD
activity is markedly augmented in cells following stable transduction with CD-
encoding retroparticles.
It has been established that CD gene transfer bestowed A549
cells with resistance to the growth inhibitory effect of ARA-C and dFdC, an
over 10-fold increase in. ICSO noted for both antimetabolites (Fig. 3).
Proliferation of CD-IRES-EGFP modified cells was considerably less
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-24-
inhibited by ARA-C and dFdC at concentrations that exhibited a pronounced
cytotoxic effect on control IRES-EGFP altered A549 cells. In addition,
selection of CD gene modified cells with ARA-C succeeding transduction
further elevated their survival. These findings indicate that ARA-C treatment
of cells ensuing transduction serves not only to eradicate the non-altered
cells but also the low CD-expressers, thusly enriching the proportion of cells
with the superior CD expression and consequently stronger drug resistance
phenotype.
The ability to successfully genetically engineer and confer
cytosine nucleoside analogue drug resistance -to primary human cells is
demonstrated, specifically in primary human lymphocytes which are
transduced with CD-containing retrovirions and cultured in vitro in various
concentrations of ARA-C. This is the first study to show CD gene transfer in
primary human cells. Survival of CD-IRES-EGFP gene-modified
lymphocytes is not significantly affected by ARA-C, even at a high dose
which entirely suppressed cell survival by control IRES-EGFP transduced
lymphocytes (Fig. 6). Therefore, the introduction of the CD cDNA into the
drug sensitive primary human lymphocytes rendered these lymphocytes
drug resistant and thus capable of surviving and even expanding in media
comprising ARA-C concentrations that usually eradicate drug sensitive cells.
These results propose the potential of the human CD gene for the positive
selection of genetically engineered human lymphocytes, which may be a
valuable tool in numerous cell and gene therapy studies. The transfer of
therapeutic genes into lymphocytes may be used in the treatment of cancer,
graft versus host disease (GVHD), AIDS, and autoimmune diseases.
Enhancing with the CD gene the population of transduced lymphocytes
expressing a second non-selectable therapeutic gene may be beneficial,
particularly in situations where poor transfer efficiency into lymphocytes is
the limiting factor for successful gene therapy applications. For instance, ex
vivo enrichment of gene-modified lymphocytes may be useful for the therapy
of patients with adenosine deaminase deficiency and for the modulation of
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-25-
GVHD following hematopoietic cell transplantation for leukemia and
lymphoma. Specifically, for controlling ~GVHD, the ex vivo enrichment using
CD of donor T lymphocytes co-expressing the HSV-TK suicide gene may
allow in vivo elimination with gancyclovir of all alloreactive donor cells
following allogeneic bone marrow transplantation. Mouse marrow stromal
cells transduced with CD-IRES-EGFP retroparticles and subsequently
combined with untransduced stroma so as to constitute 10, 20, 30, and 100%
of total cells were in all cases enriched with 2.5~,M ARA-C treatment to
>99% gene-modified populations, as assessed by flow cytori~etry analysis
for the linked GFP expression (Fig. 7). It was established that the selective
in
vitro expansion of transgene expressing cells was drug dose dependent, as
observed by exposing the 10% CD-IRES-EGFP-containing stromal cells to a
range of ARA-C concentrations (Fig. 6). It has been noted not only the dose-
dependent enrichment of the retrovector positive stromal cells but also the
dose-dependent enrichment of the cells possessing the strongest levels of
transgene expression. Accordingly, as illustrated in Fig. 9, ARA-C dose
escalation for in vitro selection of gene-modified stromal cells brought about
expression augmentation, as evidenced by GFP fluorescence increments.
Even though a high percentage of CD-IRES-EGFP positive cells were
attained with the use of moderate drug concentrations, the more powerful
doses of ARA-C selected the cells with the superior CD expression, thus
strongest drug resistance phenotype. Amplification of the CD proviral DNA
has also occurred with intensive ARA-C treatment of CD-IRES-EGFP
transduced marrow stromal cells. Characteristics of an optimal drug
resistance gene as a potential positive selectable marker comprise its
effectiveness at imparting significant drug resistance,' its small cDNA not
imposing size constraints on the therapeutic gene included in the bicistronic
vector, and very importantly its inability to raise an immune response.
Studies have demonstrated that genetically engineered primary
cells, such as lymphocytes and hematopoietic cells can be dominantly
selected not only through the concomitant expression of a drug resistance
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-26-
gene but also of a cell-surface reporter. Human and mouse hematopoietic
cells have been gene-modified to express the cell surface protein human
CD24 and were subsequently selected by fluorescence activated cell sorting
(FACS). Retroviral gene transfer of the murine cell surface marker Heat
Stable Antigen (HSA) or a truncated variant of human CD34 into human
cells has also allowed the enrichment of gene-modified cells through FACS
or immunoaffinity columns, respectively. In clinical applications it is very
likely that the expression of a non-human selectable marker by transduced
human cells will raise an immune reaction. Another selectable protein is the
human low-affinity nerve growth factor receptor for which cell sorting is
utilised to enrich genetically engineered cells based on cell-surface marking.
However, with cell surface reporters the possibility exists that untransduced
cells may be co-selected with gene-modified cells due to intercellular
exchange of proteins on the surface of cells. This obstacle may be
overcome with the use of cytoplasmic reporter proteins such as the green
fluorescent protein (GFP). Many investigators have utilised the GFP reporter
for the in vitro selection through FRCS of transfected or transduced cells
expressing high degree of fluorescence. Nevertheless, FACS may impose
significant physical stress on gene-modified cells that may be detrimental
particularly to sorted primary cells.
A further advantage of the human CD gene as a positive
selectable marker is that the drugs that it confers resistance to, cytosine
nucleoside analogues, and which are required for ex vivo selection, are
antimetabolites that will not cause DNA damage in CD-enriched cells.
While the invention has been described in connection with
specific embodiments thereof, it will be understood that it is capable of
further modifications and this application is intended to cover any varia-
tions, uses, or adaptations of the invention following, in general, the
principles of the invention and including such departures from the present
disclosure as come within known or customary practice within the art to
which the invention pertains and as may be applied to the essential
CA 02434387 2003-07-10
WO 02/057472 PCT/CA02/00042
-27-
features hereinbefore set forth, and as follows in the scope of the
appended claims.
