Gene Therapy Program, Institute of Human Genetics, Department of
Genetics, Cell Biology, and Development (C.L.S., M.D.D., J.L.F.,
R.S.M.); Department of Laboratory Medicine and Pathology (R.G.,
R.S.M.); and Department of Medicine and the Stem Cell Institute
(C.M.V.), University of Minnesota, Minneapolis, Minnesota
Expression of drug-resistant forms of dihydrofolate reductase (DHFR) in
hematopoietic cells confers substantial resistance of animals to
antifolate administration. In this study, we tested whether the
chemoprotection conferred by expression of the tyrosine-22 variant DHFR
could be used for more effective therapy of the 32Dp210 murine model of
chronic myeloid leukemia (CML). 32Dp210 tumor cells were found to be
sensitive to methotrexate (MTX) in vitro, whereas cells expressing the
tyrosine-22 DHFR gene were protected from MTX at up to micromolar
concentrations. MTX administered at low dose (2 mg/kg/day) did not
protect normal C3H-He/J mice from 32Dp210 tumor infused intravenously,
with drug toxicity limiting the administration of higher doses. Animals
engrafted with transgenic tyrosine-22 DHFR marrow were protected from
greater MTX doses (up to 6 mg/kg/day). However, the increased doses of
MTX afforded by drug-resistance gene expression surprisingly resulted
in decreased survival of the transplanted tumor-bearing animals, with
increased levels of tumor detected in peripheral blood. This apparent
exacerbation of tumor progression by MTX was not observed in DHFR
transgenic mice in which all cells and tissues contain the
drug-resistance gene. This suggests that increased tumor progression in
MTX-administered animals resulted from MTX sensitivity of a
nonhematopoietic host component, thus allowing tumor expansion. We
conclude that MTX exacerbates tumor progression in the 32Dp210 model of
CML, and that based on this model alternate DHFR inhibitors combined
with drug-resistant DHFR or other chemotherapeutic
agent/drug-resistance gene combinations may be required for the
application of drug-resistance gene expression to the treatment of CML.
 |
Introduction |
Chronic
myeloid leukemia (CML) is a disease of the hematopoietic stem cell,
resulting in the expansion and premature circulation of relatively
mature myelocytes. In the majority of cases, it is characterized by a
translocation between chromosomes 22 and 9, resulting in fusion of the
bcr and abl genes (Rowley, 1990
). The resulting bcr-abl fusion oncogene
has been shown to be necessary and sufficient for malignant
transformation of hematopoietic cells (Daley et al., 1990).
Chemotherapy for CML typically involves interferon-
, which can
prolong survival but is not curative (Hehlmann et al., 2000; Hochhaus
et al., 2000), although recently the Abl tyrosine kinase inhibitor
STI571 has shown promise in clinical trials (Druker et al., 2001a,b).
The most effective treatment for CML has been allogeneic bone marrow
transplant, but is limited by donor incompatibility leading to graft
rejection or graft-versus-host disease. Autologous marrow transplants
for CML generally lead to relapse, either due to contamination of the
graft with bcr-abl+ cells or incomplete ablation
of bcr-abl+ cells remaining in the patient
(Deisseroth et al., 1994).
Antifolates such as methotrexate (MTX) inhibit the enzyme dihydrofolate
reductase (DHFR), resulting in depletion of reduced folates necessary
for thymidylate and purine synthesis and toxicity for actively dividing
cells (Jolivet et al., 1983). Variant forms of DHFR resulting from
amino acid substitutions at various positions in the coding sequence
are less susceptible to MTX than wild-type DHFR (Simonsen and Levinson,
1983; Morris and McIvor, 1994; McIvor, 2002), and can render cells
resistant to antifolates. Introduction of a variant DHFR gene into
mouse bone marrow has been shown to confer increased antifolate
resistance to transplanted mice (Williams et al., 1987; Corey et al.,
1990; May et al., 1995; Spencer et al., 1996; James et al., 1997). The
chemoprotection provided by drug-resistant marrow potentially allows
for improved antitumor chemotherapy at greater antifolate doses (Zhao
et al., 1997b). A gene therapy approach involving transduction of
marrow with a drug-resistant DHFR gene may also improve the efficacy of
autologous marrow transplant for treatment of CML, by allowing
post-transplant antifolate selection against untransduced normal and
leukemic cells (Zhao et al., 1997a).
Many murine models of bcr-abl+ CML often result
in syndromes more closely resembling acute leukemia or lymphoma in a
subset of tumor-bearing animals (Daley et al., 1990; Voncken et al., 1995), rather than chronic-phase myeloid leukemia. One such model of
CML is the 32Dp210 cell line (Carlesso et al., 1994), established by
insertion and expression of human bcr-abl cDNA in the murine myeloid
cell line 32D (Greenberger et al., 1983). We have observed that C3H
mice infused with 32Dp210 cells exhibit rapid infiltration of
myeloblastic 32Dp210 cells in a variety of tissues, in a syndrome resembling blast-phase or acute myeloid leukemia. This article focuses
on the effect of MTX on progression of the 32Dp210 tumor model in
normal mice and in mice protected from MTX toxicity by engraftment with
marrow transgenic for the tyrosine-22 drug-resistant DHFR variant.
Although in vitro assays revealed that 32Dp210 cells are susceptible to
MTX toxicity, we found that administration of MTX to tumor-bearing mice
exacerbated progression of 32Dp210 tumor, rather than providing
improved antitumor chemotherapy. These results suggest the use of other
DHFR inhibitors in combination with drug-resistant DHFR or other
combinations of drugs/resistance genes for improved chemotherapy of
CML.
| |
Materials and Methods |
Mammalian Cell Lines and Culture.
The C3H mouse-derived,
bcr-abl+ myeloblast cell line 32Dp210 (Carlesso
et al., 1994) and all derivatives (see below) were maintained in
RPMI-1640 medium (Invitrogen, Carlsbad, CA). The
GFP+ retrovirus-producer cell line PA317-LNChRG
(Muldoon et al., 1997) was maintained in Dulbecco's modified Eagle's
medium (Invitrogen). 32Dp210+LasBD cells (Zhao et al., 1997a)
were supplemented with 20% 3T3-MoNTIL3-conditioned medium as an IL-3
source (Orchard et al., 1993). All media were supplemented with 10%
newborn calf serum (Summit Biotechnology, Fort Collins, CO), 2 mM
glutamine (Sigma-Aldrich, St. Louis, MO), 50 U/ml penicillin, 50 µg/ml streptomycin, and 0.125 µg/ml fungizone (Invitrogen). All
cells were maintained in a humidified atmosphere at 37°C and 5%
CO2.
32Dp210+GFP cells were derived by transduction of 32Dp210 cells with
LNChRG retrovirus (containing an enhanced GFP gene) at a multiplicity
of infection of approximately 9 in the presence of 8 µg/ml polybrene
(Sigma-Aldrich). Cells were exposed to virus overnight, cultured for 5 days, and GFP+ cells were isolated by
fluorescence-activated cell sorting (FACS; see below). The
FACS-isolated fluorescent cells were injected into C3H-He/J mice (see
below), recovered from the spleen of a tumor-bearing animal, and
re-sorted by FACS to obtain a fluorescent, tumorigenic cell population.
Cell Viability Assay.
Cell viability assays were conducted
using the CellTiter 96 AQueous Nonradioactive
Cell Proliferation Assay (Promega, Madison, WI) according to the
manufacturer's instructions. Briefly, 1000 cells in the appropriate
medium were inoculated into flat-bottom 96-well cell culture plates
(Corning Glassworks, Corning, NY). Varying amounts of methotrexate
(amethopterin; Sigma-Aldrich) were added to a final volume of 100 µl.
After 4 days of incubation, 20 µl of a tetrazolium indicator solution
consisting of
3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium and phenazine methosulfate was added to each well and incubated with
the cells for 2 h. Absorbance at 492 nm was then determined using
a Bio-Rad 2550 enzyme immunoassay plate reader (Bio-Rad, Hercules, CA).
Cell viability is expressed as a percentage of that observed for
control cells in the absence of MTX. Results are reported as the mean
of three determinations.
Animals and Bone Marrow Transplant.
C3H-He/J and FVB/N mice
were obtained at 6 to 8 weeks of age from the National Institutes of
Health Facility at Frederick, MD. The tyrosine-22 DHFR transgenic FVB/N
mice (line 11) used for these experiments were previous described
(James et al., 1997). F1 offspring of C3H × FVB/N matings were
designated C3F-F1 mice; line 11 DHFR transgenic FVB/N mice were used in
mating pairs with C3H-He/J mice to generate tyrosine-22 DHFR transgenic
C3F-F1 mice. Animals were housed in an Association for Assessment and
Accreditation of Laboratory Animal Care-accredited conventional
facility according to institutional guidelines.
For bone marrow transplant, transgenic tyrosine-22 DHFR marrow was
obtained by flushing marrow from femurs and tibiae of euthanized transgenic C3F-F1 mice. Transgenic marrow cells (5-10 × 106) were injected by tail vein into lethally
irradiated (800 rads cesium) nontransgenic C3F-F1 mice. Animals were
allowed to engraft for 2 months before subsequent administration of
tumor or MTX.
Tumor Administration and Therapy.
32Dp210 or 32Dp210+GFP
tumor cells were injected in 0.5 ml of Dulbecco's phosphate-buffered
saline (PBS; Celox Laboratories, Inc., St. Paul, MN) via tail vein into
C3H-He/J or C3F-F1 mice. Starting 1 day after tumor injection, PBS or
MTX was administered daily by intraperitoneal injection. Animal weight
and health were monitored daily, and moribund animals were euthanized
by CO2 asphyxiation. At regular intervals,
peripheral blood from either the periorbital vein or saphenous vein was
collected into heparinized micro-hematocrit capillary tubes (Fisher
Scientific, Pittsburgh, PA) for hematocrit determination. Statistical
comparison of survival between different groups was conducted by using
the Kaplan-Meier product limit method (Kaplan and Meier, 1958) and
calculating the log-rank statistic (Peto and Peto, 1972).
Flow Cytometric Analysis.
Flow cytometry was used to
follow in vivo growth of 32Dp210+GFP tumor. Blood samples for flow
cytometry were transferred into tubes containing an equal volume of
heparin sodium salt solution (ICN Biomedicals, Aurora, OH). The red
cells were lysed and nucleated cells were fixed either using an
ammonium chloride solution (8.99 g NH4Cl, 1 g KHCO3, and 37 mg of tetrasodium EDTA per liter;
Sigma-Aldrich) and 0.5% paraformaldehyde (Electron Microscopy
Sciences, Fort Washington, PA), respectively, or by using Optilyse B
solution (Immunotech, Marseille, France). Samples were analyzed using
the fluorescein isothiocyanate/GFP channel on a BD Biosciences
FACSCalibur (BD Biosciences, San Jose, CA) within 96 h of
preparation to detect GFP+ tumor cells. The
background level of fluorescence intensity for flow cytometry was
determined using a sample prepared from a negative control mouse that
had not received tumor or drugs, whereas a sample of 32Dp210+GFP cells
was used as a positive control. Flow cytometric data were analyzed
using FlowJo 3.2 software (Tree Star, Inc., San Carlos, CA) to
determine the percentage of cells in each sample that exhibited GFP
fluorescence greater than background.
Histology.
Spleen, liver, kidney, femur, sternum, heart,
lung, ileum, and Peyer's patch tissue samples were collected from
moribund animals. The percentage of spleen weight was determined by
dividing the spleen weight by the animal's weight at death and
multiplying by 100. Tissue samples were fixed in 10% buffered formalin
phosphate solution (Fisher Scientific) and femur and sternum samples
were decalcified in formic acid (Fisher Scientific) overnight. Samples were then embedded in paraffin, sectioned, mounted, and stained with
hematoxylin (Fisher Scientific) and eosin (Surgipath, Grays Lake, IL).
Histopathological analysis was carried out without prior knowledge of
sample identity.
| |
Results |
MTX Sensitivity of 32Dp210 Leukemia in Vitro.
The
purpose of this study was to determine the effect of MTX on tumor
progression in the 32Dp210 murine model of CML. We first determined the
effect of MTX on 32Dp210 cell viability in vitro. 32Dp210 tumor cells
containing the human bcr-abl oncogene (Carlesso et al., 1994) were
exposed to varying amounts of MTX, as described under Materials
and Methods. The 32Dp210+LasBD cell line, generated by transducing
32Dp210 cells with LasBD (a retroviral vector expressing a tyrosine-22
variant DHFR gene plus an antisense sequence directed against the
bcr-abl breakpoint; Zhao et al., 1997a) was also assayed (Fig.
1). IL-3 was included in the medium for
32Dp210+LasBD cells, because the bcr-abl antisense sequence restores
IL-3 dependence of these cells (Zhao et al., 1997a). The 32Dp210 cell
line exhibited greatly decreased cell viability at MTX concentrations
of 30 nM or greater, whereas 32Dp210+LasBD cells containing the
MTX-resistant DHFR gene retained >50% cell viability at MTX
concentrations up to 1 µM. 32Dp210 cells are thus susceptible to MTX
toxicity at doses from which cells containing a MTX-resistance gene are
protected.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 1.
MTX concentration response for growth of 32Dp210 and
32Dp210+LasBD cells in vitro. Cell growth was tested in a 4-day growth
assay at various MTX concentrations. 32Dp210+LasBD cells were
supplemented with IL-3 as required for growth (+LasBD, IL-3). 32Dp210
cells were incubated with or without IL-3 in the indicated experimental
groups (+IL-3 or no IL-3), to control for any effect of IL-3 on cell
response to MTX. Cell viability is expressed as a percentage of that
observed for control cells of the same culture in the absence of MTX.
Experiments were performed in triplicate as described under
Materials and Methods. Standard deviations for both
32Dp210 groups were less than 6% at all MTX concentrations tested and
were less than 15% at all points for the 32Dp210+LasBD group.
|
|
MTX Does Not Improve Survival of Normal 32Dp210 Tumor-Bearing
Mice.
The effect of MTX on 32Dp210 tumor progression in C3H-He/J
mice was assayed. Animals received 105 or
106 32Dp210 tumor cells on day 0, followed by
daily injections of PBS or 2 mg/kg MTX. MTX toxicity was also
determined in control mice receiving no tumor. MTX (2 mg/kg/day) did
not improve survival of tumor-bearing animals compared with animals
receiving PBS (Fig. 2, A and B).
Additionally, daily administration of 2 mg/kg MTX was eventually toxic
for normal animals after day 40. Surprisingly, this dose of MTX
significantly decreased survival of tumor-bearing mice receiving
106 32Dp210 cells compared with tumor-bearing
mice administered PBS (p < 0.02) or tumor-free,
MTX-administered control mice (p < 0.002). Hematocrit
and weight did not differ substantially between tumor-bearing animals
and animals without tumor (Fig. 3, A and
B), suggesting that animals receiving 106 tumor
cells and MTX were not dying from MTX toxicity. The gastrointestinal atrophy normally associated with MTX toxicity was not observed in mice
administered 106 tumor cells and 2 mg/kg/day MTX,
further suggesting that decreased survival of these animals was
associated with hastened tumor progression rather than enhanced MTX
toxicity. Spleens of tumor-bearing animals were greatly increased in
size compared with normal animals (Fig. 3C), consistent with the
splenomegaly observed in human CML. Administration of 2 mg/kg/day MTX
did not restore normal spleen size.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 2.
Effect of MTX on survival of tumor-bearing C3H-He/J
mice. Kaplan-Meier plots are shown for mice that received
105 (A) or 106 (B) 32Dp210 tumor cells by
intravenous injection on day 0. Also shown in both diagrams is a
control group that received no tumor (n = 10 for
each group). MTX (2 mg/kg) or PBS was administered daily by
intraperitoneal injection.
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 3.
MTX toxicity and splenomegaly in tumor-bearing
C3H-He/J mice. Animals in each of the groups depicted in Fig. 2 were
analyzed for hematocrit (A), weight (B), and spleen weight (C).
Hematocrits (A) were determined from peripheral blood samples collected
at regular intervals, as described under Materials and
Methods. Each point is the mean hematocrit for surviving mice
from the group at that time point. Standard deviations were less than
10% for all groups. Animal weights (B) were determined daily. Each
data point is the mean weight for surviving animals of the group,
plotted at 5-day intervals. Standard deviations were within 3 g at
each time point. Spleen weights (C) were measured from harvested
spleens of moribund animals, and values were expressed as a percentage
of the weight of the mouse at death, as described under
Materials and Methods. The values shown do not include
all moribund animals from each group, because tissues from some animals
were not available at the time of death. The mean spleen weight for
each group is also shown (indicated by solid vertical line). Spleen
weight for untreated control mice receiving no tumor was approximately
0.3% of the total weight of the animal (indicated by dashed line).
|
|
Histological analysis, previously undescribed for this tumor, revealed
massive accumulations of leukemic cells in the spleen (Fig.
4, A and B), as well as leukemic
infiltrates in multiple tissues, including liver (Fig. 4C), lung (Fig.
4D), bone marrow, and peristernal skeletal muscle (data not shown).
Massive necrosis of sternal bone marrow, accompanied by leukemic
infiltrates in the peristernal muscle, was common. The leukemic cell
population included a large number of blast cells with a large
polygonal to ovoid, vesicular nucleus and one or more large
eosinophilic nucleoli. The cells had abundant lightly basophilic
cytoplasm and an occasional cell had small round eosinophilic
cytoplasmic granules. This blast cell was the predominant cell in
skeletal muscle and other nonhematopoietic sites. Cells in bone marrow and spleen tended to have more pleomorphic, often band-like nuclei that
were suggestive of early myeloid stages, confirming resemblance of the
32Dp210 tumor model to an acute or blast-phase myeloid leukemia.
Similar leukemic infiltration was found in tissues of all tumor-bearing
animals whether or not MTX was administered (data not shown). Thus, MTX
did not improve survival of tumor-bearing mice, and in fact decreased
survival was observed in mice receiving 106 tumor
cells.
View larger version (135K):
[in this window]
[in a new window]
|
Fig. 4.
Histopathological analysis of 32Dp210 tumor in
C3H-He/J mice. Representative tissue samples are shown for
tumor-bearing animals administered PBS daily by intraperitoneal
injection. All sections were prepared and stained with hematoxylin and
eosin as described under Materials and Methods. A,
spleen with diffuse leukemic infiltrate. One leukemic cell is in
mitosis (200×). B, spleen, showing dense infiltrate of leukemic cells.
Some cells (center) have elongated, band-like nuclei (400×). C, liver.
Leukemic cell in metaphase, showing eosinophilic cytoplasmic granules
(400×). D, lung. Leukemic infiltrate in alveolar septa (200×).
|
|
MTX Exacerbates 32Dp210 Tumor Progression at Higher Doses of Drug
in Mice Engrafted with DHFR Transgenic Marrow.
To
determine the effect of increased MTX doses on
32Dp210 tumor progression in mice, it was necessary to render recipient animals drug-resistant by expressing a MTX-resistance gene in the
marrow. We have previously demonstrated that this can be achieved by
transplantation of marrow from DHFR transgenic mice established on an
FVB/N inbred strain background. For this study we transplanted marrow
from transgenic tyrosine-22 DHFR C3F-F1 mice into lethally irradiated
normal C3F-F1 recipient mice (see Materials and Methods). To
better track tumor progression in the animals, 32Dp210 cells were
transduced with a retroviral vector encoding an enhanced GFP as
described under Materials and Methods. After engraftment of
the transgenic marrow, mice were given 105
32Dp210+GFP tumor cells and then administered MTX at varying doses. As
previously observed, MTX did not improve survival of tumor-bearing
animals (Fig. 5A), but rather appeared to
decrease survival, particularly for animals receiving 4 mg/kg/day MTX
(median survival time of 86 days for mice receiving PBS compared with 32 days for mice receiving 4 mg/kg/day MTX). This decrease in survival
was statistically significant based on survival data through day 80, nearly 2 months after the majority of deaths occurred in the
MTX-treated group (p < 0.05 for tumor-bearing mice
receiving 4 mg/kg/day MTX compared with mice receiving PBS), but not
statistically significant after day 81 (p = 0.07). MTX
doses up to 6 mg/kg/day were not lethal for tumor-free
transgenic-marrow recipients, whereas higher MTX doses resulted in the
death of tumor-free control animals (data not shown). These results are
consistent with our previous characterization of MTX dose response in
tumor-free FVB/N mice (May et al., 1996). Average hematocrits of
tumor-bearing animals receiving 6 mg/kg/day MTX or less remained above
35 (Fig. 5B), indicating that these mice were not suffering from
MTX-mediated hematological toxicity. Additionally, histological
analysis revealed normal intestinal structure in tumor-bearing animals
administered 2 or 4 mg/kg/day MTX (Fig. 5, C and D), whereas severe
intestinal necrosis was observed in animals administered 8 or 10 mg/kg/day MTX (Fig. 5, E and F). Increased MTX-mediated
gastrointestinal toxicity was thus not observed in tumor-bearing
animals, and taken together with the hematocrit data this indicates
that the decreased survival in tumor-bearing animals observed at some
MTX doses was not caused by enhanced MTX toxicity.
View larger version (104K):
[in this window]
[in a new window]
|
Fig. 5.
Effect of MTX on tumor-bearing C3F-F1 mice
transplanted with transgenic tyrosine-22 DHFR marrow. A, Kaplan-Meier
plot of animal survival. Mice received 105 32Dp210+GFP
tumor cells by intravenous injection on day 0 (n = 8 for each group). PBS or MTX at various doses was administered daily
by intraperitoneal injection. B, average hematocrits from samples
collected at biweekly intervals. Each point is the mean hematocrit for
surviving mice from the indicated group at that time point. Standard
deviations were less than 10 at each point. Numbers refer to the MTX
dose (mg/kg) administered. C and D, histopathological analysis of small
intestines of mice receiving 2 or 4 mg/kg/day MTX. The normal mucosa
shows tall villi lined by tall columnar epithelium (125×). E and F,
histopathological analysis of small intestines of mice receiving 8 or
10 mg/kg/day MTX. Necrosis of epithelial cells has resulted in severe
mucosal atrophy with loss of villi. Attempted epithelial regeneration
is evidenced by cuboidalization and large size of luminal epithelial
cells (125×). At 10 mg/kg/day (F) the lamina propria contains an
infiltrate of mononuclear inflammatory cells. All sections were
prepared and stained with hematoxylin and eosin as described under
Materials and Methods.
|
|
Tumor load in each animal was measured by flow cytometry to detect
fluorescent GFP+ tumor cells in peripheral blood
samples collected on day 26 or 27 of the experiment (Fig.
6A). This time point corresponded
approximately to the beginning of animal mortality for those receiving
8 mg/kg/day MTX. PBS-administered animals contained only background
levels of GFP+ cells in their peripheral blood.
In contrast, the groups administered 8 mg/kg/day MTX included a
substantial number of animals exhibiting detectable (>1%)
GFP+ tumor in the peripheral blood, whereas
animals receiving 10 mg/kg/day MTX succumbed to drug toxicity before
tumor progression was observed. MTX administration (particularly a dose
of 4 mg/kg/day) thus resulted in increased numbers of circulating
32Dp210+GFP cells in the peripheral blood of tumor-bearing animals. The
lack of substantial amounts of GFP+ tumor cells
in samples from the PBS group at this time point is not anomalous,
because we have observed that 32Dp210+GFP tumor cells are generally
only detectable by flow cytometry at levels >1% within a week or less
of subsequent mortality. Spleen measurements of moribund animals
indicated that MTX did not restore normal spleen size to tumor-bearing
animals (Fig. 6B), except at 10 mg/kg/day, where animals succumbed to
drug toxicity with no tumor progression observed. Together, these data
indicate that MTX exacerbated tumor progression even at higher doses of
drug (4-6 mg/kg/day) against which animals are protected by prior
transplantation with transgenic tyrosine-22 DHFR marrow.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 6.
Effect of MTX on tumor progression in C3F-F1 mice
transplanted with transgenic tyrosine-22 DHFR marrow. A, 32Dp210+GFP
tumor cells detected in peripheral blood samples collected from mice at
day 26 or 27 of drug treatment. Blood samples were treated, and
GFP+ tumor cells were quantitated by flow cytometry as
described under Materials and Methods. Values shown are
the GFP+ cells expressed as a percentage of the total
number of cells analyzed (approximately 5000 cells total for each
sample) from individual mice. The mean value for each group is also
shown (indicated by solid vertical line). A GFP+ cell
percentage of 1% or greater (indicated by dashed line) was deemed
substantial, based on background levels observed in control samples
lacking GFP+ cells. B, spleen weights from moribund animals
expressed as a percentage of total animal weight. The values shown do
not include all moribund animals from each group, because tissues from
some animals were not available. The mean spleen weight for each group
is also shown (indicated by solid vertical line). Spleen weight for
untreated control mice receiving no tumor was approximately 0.3% of
the total animal weight (indicated by dashed line).
|
|
MTX Does Not Exacerbate 32Dp210 Tumor Progression in DHFR
Transgenic Mice.
Mice transplanted with DHFR transgenic marrow
preconditioned by lethal total body irradiation subsequently become
engrafted with predominantly donor-derived drug-resistant hematopoietic cells (May et al., 1995; James et al., 1997). Therefore, we
hypothesized that MTX exacerbation of 32Dp210 tumorigenicity resulted
from MTX sensitivity of some nonhematopoietic host tissue function. To
test this hypothesis, 4 mg/kg MTX was administered daily to transgenic
tyrosine-22 DHFR C3F-F1 mice (containing the drug-resistant DHFR gene
in all cells) after injection of 106
32Dp210+GFP cells (Fig. 7). Tumor-bearing
C3F-F1 mice lacking the tyrosine-22 DHFR transgene were administered
PBS or a reduced dose of MTX (2 mg/kg/day). In this experiment,
decreased survival was not observed in MTX-administered transgenic
animals compared with normal animals administered PBS, although neither
did MTX significantly improve survival of tumor-bearing transgenic
animals. Flow cytometry of peripheral blood samples collected on day 16 or 17 (corresponding to the onset of animal mortality in this experiment) indicated that administration of 2 mg/kg/day MTX resulted in increased levels of GFP+ tumor cells in normal
mice. In contrast, GFP+ tumor cells were
undetectable at this time point in DHFR+
transgenic mice administered 4 mg/kg/day MTX (Fig.
8A). Hematocrits were decreased at this
time point in normal animals receiving 2 mg/kg/day MTX, but were
unaffected by 4 mg/kg/day MTX administration in the transgenic group,
indicating that transgenic animals were not suffering from
hematological toxicity (Fig. 8B). These results are in direct
distinction from the results obtained in animals transplanted with DHFR
transgenic marrow, in which a similar chemoprotective effect was
observed (Fig. 5B; hematocrit), but in which GFP+
tumor cells were detected in the peripheral blood of MTX-administered animals (Fig. 6A). Measurements from moribund animals indicated that
MTX did not restore normal spleen size to tumor-bearing animals (Fig.
8C). Overall, the results from this experiment indicate that 4 mg/kg/day MTX did not exacerbate 32Dp210+GFP tumor progression in
transgenic tyrosine-22 DHFR C3F-F1 mice, in contrast to our previous
results obtained at the same MTX dose in normal C3F-F1 mice engrafted
with transgenic tyrosine-22 DHFR marrow. Tumor exacerbation in DHFR
transgenic-marrow transplant recipients was therefore most likely the
result of some MTX-sensitive nonhematopoietic host component, because
tumor exacerbation was not observed in mice containing the
MTX-resistant tyrosine-22 DHFR gene in all tissues.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 7.
Effect of MTX on survival of tumor-bearing
tyrosine-22 DHFR transgenic C3F-F1 mice. Mice received 106
32Dp210+GFP tumor cells by intravenous injection on day 0 (n = 8 for each group). Normal C3F-F1 mice received
either PBS or 2 mg/kg MTX, and transgenic C3F-F1 mice received 4 mg/kg
MTX administered daily by intraperitoneal injection.
|
|
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 8.
Effect of MTX on tumor progression in normal and
tyrosine-22 DHFR transgenic mice. A, 32Dp210+GFP tumor cells detected
in peripheral blood samples collected from mice at day 16 or 17 of drug
treatment. Values shown are the GFP+ cells expressed as a
percentage of the total number of cells analyzed (7,000-10,000 cells
total for each sample) from individual mice. The mean value for each
group is also shown (indicated by solid vertical line). A
GFP+ cell percentage of 1% or greater (indicated by dashed
line) was deemed substantial, based on background levels detected in
control samples lacking GFP+ cells. B, hematocrits of blood
samples taken at day 16 or 17 of drug treatment. Each point is the
hematocrit for an individual mouse. The mean hematocrit for each group
is also shown (indicated by solid vertical line). A hematocrit level of
35 or greater was deemed normal (indicated by dashed line), based on
samples from untreated control animals. C, spleen weights from moribund
animals, expressed as a percentage of the total animal weight. The
values shown include all moribund animals from each group. The mean
spleen weight for each group is also shown (indicated by solid vertical
line). Spleen weight for untreated control mice receiving no tumor was
approximately 0.3% of the total animal weight (indicated by dashed
line).
|
|
| |
Discussion |
The effectiveness of MTX as an antitumor agent was assayed to
evaluate a drug-resistance gene therapy approach in the 32Dp210 murine
model of chronic myeloid leukemia. MTX was found to be toxic for
32Dp210 cells in vitro at doses against which cells containing a
MTX-resistance gene were protected. In vivo studies in normal animals
revealed that MTX administration did not increase survival of
tumor-bearing animals. MTX administration at higher doses afforded by
engraftment with DHFR transgenic marrow did not reduce 32Dp210
tumorigenicity in normal transplanted animals, but rather exacerbated
tumor progression, resulting in earlier death and earlier appearance of
tumor cells in peripheral blood. This exacerbation effect was not
evident in tumor-bearing transgenic tyrosine-22 DHFR mice that received
MTX, suggesting that the promotion of tumor progression in bone marrow
transplant recipients was due to MTX sensitivity of a nontransgenic
host component. This possibility does not rule out immune suppression
in the transgenic bone marrow transplant recipients, which could occur
through disruption of a nontransgenic host component required for
immune cell maturation.
Currently, allogenic marrow transplant is the only curative treatment
for CML. Unfortunately, availability of suitable donor material is an
issue, and graft-versus-host disease has been observed in up to 68% of
CML patients transplanted with human leukocyte antigen-matched
sibling donor material (Weisdorf et al., 1991). Chemotherapeutic
approaches for CML have been only partially successful, consisting
primarily of interferon-, which can induce hematological remission
and prolong survival but is not curative (Hehlmann et al., 2000;
Hochhaus et al., 2000). However, STI571, a selective inhibitor of Abl
tyrosine kinase, has proved effective in inhibiting growth of
bcr-abl+ cells (Druker et al., 1996), and recent
clinical trials have shown promise for the treatment of CML (Druker et
al., 2001a,b). MTX has been effective in treatment of a variety of
malignancies, including choriocarcinoma, breast cancer, non-Hodgkin's
lymphoma, and acute lymphoid leukemia (Jolivet et al., 1983; Bertino,
1993). MTX is commonly used to control graft-versus-host disease after allogeneic marrow transplant (Simpson, 2000). MTX is not typically used
as an antitumor agent for treatment of CML, although MTX coadministration has been shown to improve the efficacy of
interferon- in delaying CML progression to blast crisis (Kanda et
al., 1999).
Numerous studies have reported the protection of mice from toxic doses
of antifolate (MTX or trimetrexate) by expression of drug-resistant
DHFR in hematopoietic cells (Williams et al., 1987; Corey et al., 1990;
Li et al., 1994; May et al., 1995; Spencer et al., 1996; James et al.,
1997). Administration of higher doses of antifolate in animals thus
protected could allow for more effective treatment of tumors known to
be sensitive to antifolates (Zhao et al., 1997b). Drug-resistant DHFR
expression could also be used to allow antifolate administration at
higher doses, which may be effective against tumors that are not
usually sensitive to antifolates at lower doses, such as CML.
Autologous marrow transplants for treatment of CML commonly result in
relapse either due to bcr-abl+ tumor cells
contaminating the graft, or incomplete elimination of tumor cells in
the host during the preparative regimen for bone marrow transplant
(Deisseroth et al., 1994). Introduction of a drug-resistance gene into
autologous marrow before transplant could allow for post-transplant
drug administration to selectively ablate leukemic cells lacking the
drug-resistance gene. One problem with this approach is the possibility
of introducing the drug-resistance gene into tumor cells contaminating
the donor marrow. This problem could be addressed for CML by including
antisense sequences directed against the fusion region of the bcr-abl
message, thus restoring a more normal phenotype to transduced
bcr-abl+ cells in the graft. This strategy was
previously examined using the 32Dp210 model of CML, demonstrating a 3 log-fold reduction in tumorigenicity for tumor cells transduced with
LasBD, a retroviral vector containing the tyrosine-22 DHFR gene in
addition to antisense sequences directed against the bcr-abl breakpoint
(Zhao et al., 1997a).
This article examines the effectiveness of
expressing the variant tyrosine-22 DHFR gene in mouse marrow for
improved MTX chemotherapy in the 32Dp210 model of CML. As with previous
studies involving antifolate administration in other mouse strains
transplanted with either transgenic or retrovirally transduced marrow
containing a variant DHFR gene (Williams et al., 1987; Corey et al.,
1990; Li et al., 1994; Zhao et al., 1994, 1997b; May et al., 1995,
1996; Spencer et al., 1996; James et al., 1997, 2000), we found that the tyrosine-22 DHFR gene conferred increased resistance of C3F-F1 mice
to MTX, allowing for MTX administration at higher doses. However, the
increased MTX administration afforded by engraftment with
drug-resistant transgenic marrow did not improve animal survival in the
32Dp210 model of CML, but rather accelerated progression of the tumor.
The tyrosine-22 DHFR transgenic mouse system has also been assayed for
improved chemotherapy of FMC, a mammary adenocarcinoma established in
this laboratory from FVB/N mice. MTX administration was found to be
relatively ineffective at inhibiting FMC tumor progression in animals
transplanted with drug-resistant marrow, but MTX did not exacerbate FMC
tumor progression in those animals (J. L. Frandsen, B. Weigel,
C. L. Sweeney, M. D. Diers, R. Gunther, and R. S. McIvor, manuscript in preparation). Therefore, the accelerated
progression of 32Dp210 tumor by MTX is not characteristic of all tumors
in this transgenic system.
The fact that MTX did not increase 32Dp210 tumor progression in DHFR
transgenic animals (i.e., animals in which all tissues contain the DHFR gene) provides some insight into the mechanism by
which MTX exacerbates tumorigenicity in normal and bone marrow transplant recipient animals. The results are consistent with the presence of some MTX-sensitive nonhematopoietic host function that
provides an appropriate setting for tumor growth upon MTX administration. MTX administration could thus create a niche, or some
space, in this drug-sensitive tissue, which allows the tumor to become
established, grow, and expand. Alternately, given the apparent role of
the immune response in controlling human CML (Molldrem et al., 2000; Wu
et al., 2000), tumor exacerbation in this model may be due to some
degree of immune suppression by MTX. However, the lack of MTX-mediated
acceleration of tumor progression in wholly transgenic animals argues
against a marrow-derived, MTX-sensitive host component. This
interpretation would imply that transgenic immune cells (e.g., T or B
cells) retain some sensitivity to MTX despite the presence and
expression of the drug-resistant DHFR gene.
Whatever the mechanism by which MTX exacerbates tumorigenicity, it is
apparent that the combination of DHFR gene transfer with increased
doses of MTX is an ineffective approach in preclinical studies with the
32Dp210 model of CML. Some modification of this strategy will be
necessary to achieve an efficacious response. It is possible that the
application of DHFR gene transfer for CML in this model could be
improved by the use of other DHFR inhibitors besides MTX. In
particular, the lipophilic antifolate trimetrexate is also a potent
inhibitor of DHFR, but does not rely on the reduced folate carrier for
transport into cells and does not require polyglutamylation for
inhibition of DHFR (Lin and Bertino, 1987). Trimetrexate is a more
specific inhibitor of DHFR than MTX, because MTX-polyglutamates also
directly inhibit thymidylate synthase as well as glycinamide ribonucleotide and aminoimidazole carboxamide ribonucleotide
transformylases involved in de novo purine synthesis (Takimoto, 1996).
Additionally, transport of exogenous nucleosides can potentially rescue
cells from MTX toxicity (Nelson and Drake, 1984; Sur et al., 1993). Use
of a nucleoside transport inhibitor can restore MTX toxicity to cells
expressing wild-type DHFR while maintaining differential toxicity
relative to cells expressing drug-resistant DHFR (Warlick et al.,
2000), and may allow for increased effectiveness of antifolates for
chemotherapy in the 32Dp210 murine model of CML. Trimetrexate in
combination with nucleoside transport inhibition has been used effectively for in vivo selection of murine hematopoietic stem cells
(Allay et al., 1998). Alternatively, the introduction and expression of
other drug-resistance genes, such as the multidrug resistance gene 1 (Aran et al., 1999) or
O6-methylguanine-DNA-methyltransferease
gene (Reese et al., 1996) could be used to provide chemoprotection to
recipient animals and may prove beneficial for improved tumor
chemotherapy. These approaches are currently under investigation for
ultimate application in the treatment of CML in combination with
antisense sequences directed against the bcr-abl oncogene, as
previously described (Zhao et al., 1997a).
We acknowledge the assistance of the Flow Cytometry Core
Facility of the University of Minnesota Cancer Center, a comprehensive cancer center designated by the National Cancer Institute, supported in
part by P30 CA77598.
This study was supported by research Grants CA74887 and CA60803
from the National Institutes of Health.
CML, chronic myeloid leukemia;
MTX, methotrexate;
DHFR, dihydrofolate reductase;
GFP, green fluorescent
protein;
IL-3, interleukin-3;
FACS, fluorescence-activated cell
sorting;
PBS, phosphate-buffered saline.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
