| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
Department of Pharmacology and Developmental Therapeutics Program, Cancer Center and Section of Comparative Medicine, Yale University School of Medicine, New Haven, Connecticut 06520
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The P-gly, expressed by the MDR1 gene in humans and two
closely related genes, mdr1a and mdr1b, in the
mouse, and MRP1 play central roles in export pump-mediated resistance
through the active extrusion of a wide range of structurally diverse
antineoplastic agents including the Vinca alkaloids, the
epipodophyllotoxins, and the anthracyclines (3, 4, 5, 6)
.
Although the P-gly transports free drugs, MRP1 can be considered to be
an organic anion transporter capable of transporting a broad spectrum
of organic anion conjugates of glutathione, glucuronic acid, and
sulfate. In addition, we have shown with etoposide (7)
and
Loe et al. (8)
with vincristine that mrp1 also
functions as a cotransporter of xenobiotics and glutathione. In keeping
with these findings, we have also demonstrated that levels of GSH in
mrp1(-/-) mice were elevated by 25–90% in most tissues,
especially in those tissues that are known to express high levels of
mrp1 (9)
. That tissue increases in GSH in
mrp1(-/-) mice were not attributable to the increased
synthesis of GSH was supported by the finding that the levels of 
-
glutamylcysteine synthase, the rate-limiting enzyme in the
synthesis of GSH, was not significantly different in any of the tissues
of mrp1(+/+) and mrp1(-/-) mice
(9)
. Prior to these findings, a variety of studies had
provided evidence that GSH was required for the transport of
chemotherapeutic agents (see Ref. 3
for appropriate
references).
Schinkel et al. (10 , 11) have derived mice deficient in the mdr1a gene [mdr1a(-/-)] and, in addition, have derived mice deficient in the mdr1b gene [mdr1b(-/-)], and in both the mdr1a and mdr1b genes [mdr1a/1b(-/-)]. All three of these gene knockout animals are normal in all measured physiological parameters, displaying normal viability, fertility, and life span, as well as normal levels of a range of serum enzymes, proteins, electrolytes, and hematological parameters.
In clinical trials, the P-gly has often been shown to be elevated in the hematological malignancies, particularly after the failure of multiple drug therapy (12) . Thus, the MDR phenotype as a cause of resistance in acute myelocytic leukemia and multiple myeloma and possibly in the late stages of non-Hodgkin’s lymphoma and acute lymphocytic leukemia has been documented. The role of MDR1 gene expression in the clinical resistance of solid tumors, however, is currently not firmly established (13) . Nonetheless, in several malignancies, such as acute myelocytic leukemia, various childhood cancers, and advanced breast cancer, overexpression of the MDR1 gene has been shown to correlate with a poor response in patients receiving cancer chemotherapeutic agents (reviewed in Ref. 14 ). In colon cancer, renal cell carcinoma, primary breast cancer, and osteosarcoma, clinical studies have shown that P-gly positivity is associated with aggressive tumor behavior and is a strong predictor of treatment outcome. Whether in these instances the P-gly is a marker for drug resistance, tumor aggressiveness, or both is currently unknown.
We (9) and Wijnholds et al. (15) have shown that disruption of mrp1 did not affect the viability or fertility of mice, nor were hematological parameters or levels of serum enzymes, proteins, and electrolytes different in mrp1(-/-) and mrp1(+/+) mice. However, mrp1(-/-) mice were hypersensitive to a relatively large number of anticancer drugs (9 , 15 , 16) . The demonstration that the lack of mrp1 in mrp1(-/-) mice led to toxicity to the oropharyngeal mucosa and the seminiferous tubules of the testis in etoposide phosphate-treated animals indicates that mrp1 protects these tissues against damage from mrp1 substrates (17) . The expression of MRP1 has been detected in a number of human cancers and shown to be associated with drug resistance or reduced patient survival in a variety of tumor types, including lung, breast, colon, and gastric cancers, as well as in the childhood cancers, neuroblastoma and retinoblastoma (18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28) .
The P-gly and MRP1 have been targets for the development of inhibitors of transport in an effort to restore sensitivity to neoplastic cells using these export pumps to extrude tumoricidal agents (reviewed in Refs. 29, 30, 31, 32 ). The identification of several agents that have the capacity to inhibit both the P-gly and MRP1 make it conceivable that new serious toxicities may also result from the use of such agents (9 , 33, 34, 35, 36, 37, 38, 39, 40) .
The present report is an effort to further understand the consequences of multiple ABC transporter gene disruption and its impact on antineoplastic agent toxicity. The findings demonstrate that the P-gly and MRP1 are compensatory transporters for vincristine and etoposide and that a functional deficiency in these transporters can produce unexpected serious new toxicities.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Western Blotting.
Western blot analyses of mrp1 and mdr1a/1b were performed on protein
obtained from cultured mouse embryo fibroblasts using the monoclonal
murine mdr1 antibody C219 (Signet, Dedham, MA) at a 1:80 dilution and
the monoclonal rat mrp1 antibody MRPr1 (Signet) at a concentration of
0.3 µg/ml. Fibroblasts were grown to near confluence in
25-cm2 flasks, harvested, and resuspended in
Tris-buffered saline [10 mM Tris-HCl (pH 7.6), 0.1 M NaCl,
and 1 mM EDTA] containing a mixture of protease inhibitors
(2 mM phenylmethylsulfonyl fluoride, 1 µg/ml of
leupeptin, and 1 µg/ml of aprotinin), lysed by adding an equal volume
of 2x SDS containing gel-loading buffer [100 mM Tris-HCl
(pH 6.8), 200 mM DTT, 4% SDS, 0.2% bromphenol blue, and
20% glycerol], placed in a boiling water bath for 5 min, and vortexed
vigorously. Extracts were separated by electrophoresis on a 10%
SDS-polyacrylamide gel and transferred onto nitrocellulose membranes.
The membranes were blocked with 5% dry milk in TBST [20
mM Tris-HCl (pH 7.6), 137 mM NaCl, and 0.01%
Tween 20] for 30 min, incubated overnight with the respective antibody
in TBST containing 5% milk, washed with three changes of TBST for a
total of 30 min, and then incubated with horseradish
peroxidase-conjugated sheep antimouse IgG or rabbit antirat IgG for
1 h and washed for 30 min with TBST (three changes). The
immunoreactive proteins were visualized by the enhanced
chemiluminescence method (Amersham Corp., Arlington Heights, IL).
Embryo Fibroblast Cultures.
Embryo fibroblast lines were prepared from wild type;
mrp1(-/-); mdr1a/1b(-/-); and
mdr1a/1b(-/-), mrp1(-/-) mice on the 13th or
14th day of pregnancy. The embryos were dissected free of associated
membranes, and placentas and were rinsed thoroughly in PBS containing
100 units/ml of penicillin and 0.1 mg/ml of streptomycin. In a new
sterile Petri dish, the embryos were finely minced in 2 ml of
trypsin/EDTA, incubated at 37°C for 30 min, and then further
disrupted by repeated aspiration in a 1-ml syringe. Fifteen ml of DMEM
containing 15% FCS were added, the suspensions were transferred to a
50-ml polypropylene tube, and the larger pieces were allowed to settle
for 1–2 min. The supernatants were then transferred to clean tubes and
centrifuged at 600 x g for 5 min, and the
resulting pellet was resuspended in 7.5 ml of DMEM containing 15% FCS,
plated in 25-cm2 tissue culture flasks, and
incubated at 37°C with 5% CO2. Cells typically
grew to confluency after 
48 h with a change in medium at 24 h
to remove cellular debris.
In Vitro Drug Sensitivities of Embryo Fibroblasts.
Embryo fibroblasts were grown from frozen stocks and grown as passage 0
in DMEM supplemented with 15% FCS. The cultures were subsequently
passaged every 3–4 days, and toxicity experiments were performed on
cells up to passage four. Sensitivities to vincristine, etoposide, and
paclitaxel (Biomol Plymouth Meeting, PA) were measured in culture using
the CellTiter 96 Cell Proliferation Assay (Promega Corp., Madison, WI).
Briefly, confluent embryonic fibroblasts were seeded in 96-well plates
at 3000 cells/well in complete DMEM and incubated for 24 h, after
which drug was added such that the final volume of medium/well was 100
µl. After 96 h of exposure to either vincristine or paxlitaxel
or 72 h to etoposide, 20 µl of MTS tetrazolium solution
(Promega) were added, and incubations were continued for an additional
2 h at 37°C. The soluble MTS tetrazolium formazan product formed
by the dehydrogenase activity of viable cells was quantified at 490 nm
using a microplate reader, and the sensitivity to drugs was expressed
as a percentage of untreated controls.
In Vivo Toxicity of Vincristine and Etoposide.
In vivo etoposide phosphate (generously provided by the
Bristol-Myers Squibb Co., Princeton, NJ) toxicity using male mice and
vincristine toxicity using female mice were determined as follows.
Groups of three to six wild-type; mrp1(-/-);
mdr1a/1b(-/-); and mdr1a/1b(-/-),
mrp1(-/-) male mice were treated with single i.p. doses of
etoposide phosphate ranging from 50 to 200 mg/kg of body weight. In a
similar manner, groups of three or six wild-type;
mrp1(-/-); mdr1a/1b(-/-); and
mdr1a/1b(-/-), mrp1(-/-) female mice were
treated with single i.p. doses of vincristine ranging from
0.03125 to 10 mg/kg of body weight. The survival of treated
animals was followed for 30 days. In both cases, six mice were used in
treatment groups that were one dose above and below the MTD. The MTD
was expressed as the highest dose of drug at which all of the animals
of a given genotype survived. The sensitivity ratio was calculated as
the ratio of the MTD for wild-type animals divided by the MTD for each
respective genotype and was expressed as a hypersensitivity index.
Peripheral WBC Counts after Vincristine Administration.
Peripheral WBCs of wild-type; mrp1(-/-);
mdr1a/1b(-/-); and mdr1a/1b(-/-),
mrp1(-/-) mice given a single i.p. injection of 1 mg/kg of
vincristine were determined at various times after treatment with the
Vinca alkaloid using a Coulter Multisizer II.
Necropsy.
Necropsy, including histological evaluation, was performed on female
mice, including untreated control and drug-treated animals from each of
the four genotypes [wild-type; mrp1(-/-);
mdr1a/1b(-/-); and mdr1a/1b(-/-),
mrp1(-/-)] 3 days after a single i.p. injection of 2
mg/kg of vincristine. Toxicity was further evaluated in male mice 3
days after a single i.p. injection of 1 mg/kg of vincristine.
Blood Chemistries.
Peripheral blood chemistries (Antech Diagnostics, Farmingdale, NY)
including glucose, urea nitrogen, creatinine, total protein, albumin,
total bilirubin, alkaline phosphatase, alanine aminotransferase,
aspartate aminotransferase, cholesterol, calcium, phosphorous, sodium,
potassium, chloride, albumin:globulin ratio, blood urea
nitrogen:creatinine ratio, globulin, lipase, amylase, triglycerides,
creatine phosphokinase, gamma glutamyl transpeptidase, magnesium, and
calculated osmolality were measured. These values were obtained from
pools of blood collected by cardiac puncture (three mice each) of
wild-type and combined mdr1a/1b(-/-),
mrp1(-/-) male and female mice ranging in age from 7 to 9
weeks. Mice were anesthetized with sodium pentobarbitol (70 mg/kg)
prior to the collection of blood.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The presence of the 5-kb band in the absence of a 2-kb band indicates
homozygosity for the disrupted mrp1 allele. Because all of
the F1 animals were heterozygous at all three
genetic loci, some of the offspring resulting from an
F1 x F1 cross
contained disruptions in both alleles of all three genes. Generation
F2 offspring testing positive for the homozygous
disruption of mrp1 were further tested for
mdr1a/1b(-/-) status by PCR (Fig. 1B)
. Mouse
numbers 2, 6, 8, and 10 are representative animals displaying
homozygous disruption of the mrp1, mdr1a, and
mdr1b genes. Ten of 29 mrp1(-/-) mice (5
females and 5 males), evaluated for mdr1a/1b status, tested positive
for the homozygous disruption of mdr1a and mdr1b.
Male and female mice with the mrp1(-/-), mdr1a/1b(-/-)
genotype were mated to establish the triple knockout colony.
|
. Using the murine monoclonal antibody C219 to detect mdr1a/1b, the
presence of mdr1a and mdr1b was confirmed in wild-type and
mrp1(-/-) embryonic fibroblast lines and its absence
documented in mdr1a/1b(-/-) and mdr1a/1b(-/-),
mrp1(-/-) fibroblast lines. Similarly, using the rat monoclonal
antibody MRPr1, the presence of mrp1 in wild-type and
mdr1a/1b(-/-) animals was observed, and its absence in the
mrp1(-/-) and mdr1a/1(-/-), mrp1(-/-) mice
was documented.
|
. Cell viability was expressed as a percentage of control over a range
of drug concentrations and from which increased sensitivity in
comparison to wild-type cells, based upon comparative
IC50s, was calculated for each genotype. In the
case of vincristine, mrp1(-/-);
mdr1a/1b(-/-); and mdr1a/1b(-/-),
mrp1(-/-) embryonic fibroblasts exhibited increased
sensitivities of approximately 1.7-, 3.5- and 12-fold, respectively,
following exposure to the Vinca alkaloid for 96 h.
These findings were suggestive of a synergistic effect in the setting
of the triple gene deficiency. For etoposide, after a 72-h incubation,
mrp1(-/-); mdr1a/1b(-/-); and
mdr1a/1b(-/-), mrp1(-/-) animals showed
increased sensitivities of 3.1-, 1.4- and 4.3-fold, respectively. With
paclitaxel exposure for 96 h, both mdr1a/1b(-/-) and
mdr1a/1b(-/-), mrp1(-/-) cells demonstrated
an equivalent degree of enhanced sensitivity of about 25-fold over
wild-type cells, whereas mrp1(-/-) cells displayed no
increased sensitivity, a finding consistent with paclitaxel not being a
substrate for mrp1.
|
. The maximum tolerated single i.p. doses of etoposide phosphate for
wild-type; mrp1(-/-); mdr1a/1b(-/-); and
mdr1a/1b(-/-), mrp1(-/-) mice were 175, 100,
100, and 50 mg/kg, respectively. Although both the
mrp1(-/-) and mdr1a/1b(-/-) mice tolerated
the 100 mg/kg dose, at the next higher dose tested four of the six
mrp1(-/-) animals survived versus two of the
six mdr1a/1b(-/-) mice. The hypersensitivity factor for
these animals toward etoposide phosphate was 1.75 for both
mrp1(-/-) and mdr1a/1b(-/-) and 3.5 for the
triple knockout mice as illustrated in Fig. 4
. Analogous vincristine mortality data are presented in Table 2
. The maximum tolerated single i.p. doses of vincristine for wild-type;
mrp1(-/-); mdr1a/1b(-/-); and
mdr1a/1b(-/-), mrp1(-/-) mice were 4, 1, 0.25
and 0.03125 mg/kg, respectively. The hypersensitivity factors for the
various genotypes for the Vinca alkaloid were significantly
more pronounced than for etoposide [4-, 16-, and 128-fold for the
mrp1(-/-); mdr1a/1b(-/-); and
mdr1a/1b(-/-), mrp1(-/-) mice, respectively
(Fig. 5)
]. The increase in sensitivity toward vincristine observed in mice with
disruptions in the three transport genes was consistent with the
enhanced sensitivity of embryonic fibroblasts to this agent observed
in vitro.
|
|
|
|
. After an approximately 10–20% decrease in WBC counts by day 3, the
levels of WBCs in wild-type and mrp1(-/-) mice began to
recover, essentially returning to baseline by posttreatment day 4. The
nadir in the WBC count for the mdr1a/1b(-/-) mice occurred
on the second day after treatment with the Vinca alkaloid,
with values just under the 50% level, followed by a progressive
recovery in the white cell count to >70% of normal by posttreatment
day 5 for three of six surviving mice. The mdr1a/1b(-/-),
mrp1(-/-) mice showed a decrease in the WBC counts to
approximately 20–25% of normal by posttreatment day 3 and remained
depressed without evidence of recovery until the deaths of all mice.
|
In an analogous manner, histological examination of the bone marrow
after vincristine administration demonstrated enhanced toxicity with
increasing gene disruption. Thus, 72 h after a 1-mg/kg dose of the
Vinca alkaloid, changes in the bone marrow of
mrp1(-/-) mice occurred, characterized by a relatively
modest depletion of myeloid elements. Moderate depletion of myeloid
elements was also produced in mdr1a/1b(-/-) animals,
whereas triple knockout mice exhibited severe marrow destruction, with
almost complete loss of myeloid elements, frank necrosis, hemorrhage,
and fibrin deposition. Representative H&E-stained sections of bone
marrow from the femurs of untreated animals and from those treated
72 h earlier with 2 mg/kg of vincristine are shown in Fig. 7
. No significant change from that of untreated animals was observed in
vincristine-treated marrow from wild-type animals. Expansion of the
blood-filled medullary sinusoids because of the modest loss of myeloid
elements was observed in mrp1(-/-) mice, whereas a more
severe reduction of hematopoietic precursors, with many expanded
blood-filled vascular spaces and prominent stroma replacing
hematopoietic precursors, was seen in mdr1a/1b(-/-)
animals. Bone marrow from triple knockout mice exposed to the
Vinca alkaloid exhibited extensive acute necrosis and
hemorrhage with loss of almost all of the hematopoietic precursors.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
mdr1a(-/-) mice have a complete loss of the P-gly present
in the brain capillaries and the intestinal epithelium, demonstrating
that the mdr1a gene is responsible for the P-gly found in
these tissues and that the absence of the mdr1a P-gly does not lead to
the activation of mdr1b P-gly in these or in most other tissues
(11)
. However, an up-regulation of mdr1b
occurred in the liver and kidneys of mdr1a(-/-) mice,
suggesting that a compensatory mechanism was operative in these
tissues. The most striking effect of the treatment of wild-type and
mdr1a(-/-) mice with a single dose of 6 mg/kg of VBL was a
12-fold higher concentration of VBL in the brain of the
mdr1a(-/-) mice than in this tissue in wild-type animals
at 4 h after the drug, and this differential in the brain
increased with time because of slower elimination of VBL in
mdr1a(-/-) mice. The lethal dose of VBL was 3-fold
lower in mdr1a(-/-) mice than in their wild-type
counterparts, and the signs of general toxicity were similar in
parental and knockout animals.
mdr1b(-/-) mice did not exhibit any significant differences from wild-type animals in the tissue distribution of [3 H]digoxin, and this finding presumably corresponded to the compensatory use of the mdr1a gene in the major organs of mdr1b(-/-) mice (10) . In contrast, mdr1a/1b(-/-) mice exhibited increased plasma levels of [3 H]digoxin and marked increases in this agent in the brain and in the testes of mice compared with that of parental mice. In addition, [3 H]digoxin accumulation in adrenal glands and ovaries was increased over that of plasma levels and that present in most of the other tissues of the mdr1a/1b(-/-) mice compared with mdr1a/1b(+/+) mice. Studies of the elimination rate of [3 H]digoxin in mdr1a/1b(-/-) mice demonstrated that the liver has a substantial [3 H]digoxin excretion capacity that is distinct from the mdr1-type P-gly and that the urinary excretion of [3 H]digoxin was not impaired in mdr1a/1b(-/-) mice, indicating that the mdr1-type P-gly is not essential for the excretion of this agent by the kidney. In contrast, the direct elimination of [3 H]digoxin from the intestine in mdr1a/1b(-/-) animals was markedly decreased compared with mdr1a/1b(+/+) mice. Comparable findings were also demonstrated for the antineoplastic agent paclitaxel in mrp1a/1b(-/-) mice, suggesting the presence of a compensatory transporter(s) in the liver and the kidneys.
The P-gly substrate rhodamine 123 has been used to measure the importance of the P-gly to hematopoietic stem cells by determining the rate of efflux of this dye from partially purified hematopoietic progenitor cells from the bone marrow of wild-type and knockout mice (10) . The rate of rhodamine efflux from these hematopoietic cells from mdr1a(-/-) and mdr1b(-/-) mice was decreased relative to these cells from wild-type animals, whereas a supra-additive decrease was observed in these cells from mdr1a/1b(-/-) mice, demonstrating that both mdr1a and mdr1b genes contribute substantially to drug efflux from hematopoietic progenitor cells.
To ascertain whether the baseline expression of mrp1 protects mice from the toxic effects of xenobiotics, limited toxicity tests have been conducted with etoposide and etoposide phosphate (Etopophos), a water-soluble etoposide ester that is completely and rapidly dephosphorylated to etoposide in plasma (9) . Etoposide phosphate, injected i.p. as a single dose, was twice as toxic to mrp1(-/-) mice than to mrp1(+/+) mice, with calculated LD50s of 95 and 190 mg/kg, respectively. One of the main toxicities of etoposide in humans, as well as in mice, is to the bone marrow. To determine whether treatment with etoposide phosphate resulted in differential bone marrow toxicity to mrp1(+/+) and mrp1(-/-) mice, in a previous study we measured the total WBC count at different times after the i.p. injection of 150 mg/kg of etoposide phosphate (9) . After a rapid initial drop in the WBC count, a nadir was reached between days 2 and 3 in both mrp1(+/+) and mrp1(-/-) mice. Subsequently, the leukocyte count recovered in wild-type animals but not in mrp1 knockouts. This result implied that etoposide phosphate exerted a differential toxicity to the bone marrow of wild-type and mrp1 knockout mice. These findings were corroborated by a pathological examination of the bone marrow and spleen of wild-type and mrp1 knockout mice 5 days after treatment with 150 mg/kg of etoposide phosphate. Although in mrp1 knockout animals the bone marrow exhibited a severe depletion of nucleated cells and the spleen exhibited a depletion of myeloid activity in the red pulp, in wild-type animals the bone marrow and spleen were either normal or hypercellular (data not shown). In contrast, although Wijnholds et al. (15) demonstrated that mrp1(-/-) mice were hypersensitive to etoposide, expressed as increased loss of body weight and increased mortality, these authors found no difference in leukopenia and thrombocytopenia between mrp1(-/-) and mrp1(+/+) mice.
In the present report, we demonstrate that mice lacking the three genes mdr1a, mdr1b, and mrp1 develop normally and are without physical dysmorphology, internal anatomical abnormality, and known endogenous biochemical abnormality. These animals reproduce and show normal viability and, in the absence of pharmacological challenge, are indistinguishable from wild-type; mrp1(-/-); or mdr1a/1b(-/-) mice. These findings corroborated those of Wijnholds et al. (42) , who have independently developed mdr1a/1b(-/-), mrp1(-/-) mice. Retinal degeneration was observed in some of the triple knockout mice; this finding is consistent with the retinal degeneration known to arise on the genetic background of the mdr1a/1b(-/-) strain used to breed the combined mdr1a/1b(-/-), mrp1(-/-) mice (41) and therefore was assumed to be pathogenically unrelated to the combined triple gene knockout. It is provocative, however, to note that mutations in another ABC transporter (ABCR) have been associated with retinal pathology (43) .
The consequences of combined mdr1a/1b, mrp1 export pump deficiency in
the face of pharmacological challenge were evaluated with vincristine,
etoposide, and paclitaxel in vitro in embryonic fibroblasts
from triple knockout mice, as well as with vincristine and etoposide
in vivo. In vitro studies with vincristine using embryonic
fibroblasts demonstrated a picture of increasing sensitivity correlated
with increased transporter deficiency, yielding enhanced sensitivities
of 1.7-, 3.5- and 12-fold over wild-type; mrp1(-/-);
mdr1a/1b(-/-); and mdr1a/1b(-/-),
mrp1(-/-) embryonic fibroblasts, respectively. Studies
with etoposide in embryonic fibroblasts indicated that triple knockout
embryonic fibroblasts were 4.3-fold more sensitive than wild-type
fibroblasts; this hypersensitivity appeared to be accounted for
primarily by a deficiency in mrp1, with mdr1a/1b contributing little if
any to the observed increase in sensitivity to the epipodophyllotoxin.
Paclitaxel, which is preferentially transported by the P-gly, was no
more cytotoxic to embryonic fibroblasts lacking both the P-gly and mrp1
than fibroblasts deficient in the P-gly alone, with both genotypes
demonstrating an 25-fold enhanced sensitivity compared with
wild-type and mrp1(-/-) cells. These results corresponded
to those reported recently by Allen et al.
(44)
, who found increased sensitivity to vincristine and
etoposide in independently developed triple knockout fibroblast cell
lines. An unexplained finding was a 28-fold increase in sensitivity to
vincristine with only a 7.1-fold increase in sensitivity to VBL over
that of wild-type fibroblasts.
Vincristine had an even more pronounced effect in vivo, with mortality data indicating that the triple knockout mice were 128 times more sensitive to the Vinca alkaloid in terms of lethality than were wild-type animals, whereas 16- and 4-fold enhanced sensitivities were observed with mdr1a/1b(-/-) and mrp1(-/-) genotypes, respectively. The in vivo toxicity of vincristine also exhibited genotype-related trends in the time of death at a given dosage of drug, such that at 4 mg/kg of the Vinca alkaloid, all of the wild-type mice survived, whereas the average survival times for mdr1a/1b(-/-), mrp1(-/-); mdr1a/1b(-/-); and mrp1(-/-) animals were 4, 5, and 7 days, respectively. Qualitative assessment of vincristine-treated mice also demonstrated evidence of unsteady gait and tremor, which appeared to be the most severe in the triple knockout animals.
A change in peripheral WBC counts that correlated with genotype was observed after treatment with 1 mg/kg of vincristine. Thus, the WBC count in triple knockout mice dropped markedly lower than that occurring with the other genotypes and exhibited little recovery. mdr1a/1b(-/-) animals showed an intermediate response between that of the triple knockout and mrp1(-/-) mice, which was apparent between posttreatment days 2 and 5; this finding is consistent with the known expression of the P-gly in pluripotent stem cells of the bone marrow (45) . These observations were further supported by the extensive histological manifestations of bone marrow toxicity after vincristine administration to triple knockout mice. Taken together, the in vivo histology and mortality data, as well as the in vitro toxicity with the Vinca alkaloid, present a picture of enhanced drug sensitivity in animals with the combined disruption of the three transport genes, such that the functional absence of these three transporters resulted in greater than additive toxicities compared with those observed in the absence of either mdr1a/1b or mrp1 alone. This phenomenon implies that the P-gly and mrp1 are compensatory transporters in the bone marrow and intestinal mucosa, as well as in other tissues. Furthermore, the relative importance of these transporters was clearly drug and tissue dependent, as illustrated by the in vivo sensitivity to etoposide, as well as by the in vitro response of embryonic fibroblasts to paclitaxel. Thus, although the same degree of enhanced toxicity to etoposide was not observed across the transporter-deficient genotypes, as was seen with vincristine, the triple knockout mice exhibited a 3.5-fold increased sensitivity to the epipodophyllotoxin, whereas a 1.75-fold enhanced sensitivity occurred in both the mdr1a/1b(-/-) mice and mrp1(-/-) mice. The enhanced sensitivity to etoposide corresponded with that described by Wijnholds et al. (42) , who also showed increased accumulation of etoposide in the colon, brown adipose tissue, salivary gland, heart, and the female urogenital system of triple knockout mice.
The findings also permit the suggestion that the therapeutic differential toxicity of antineoplastic agents effluxed by the P-gly and/or MRP1 to malignant cells relative to normal tissues, such as the bone marrow and the gastrointestinal mucosa, may be attributable, at least in part, to the presence of ABC transporters in the normal tissues.
It is conceivable that additional transporters other than the P-gly and
mrp1 may be present in sensitive tissues such as the bone marrow and/or
the intestinal mucosa. Such a possibility may explain, at least in
part, the markedly different hypersensitivity factors seen with
vincristine and etoposide in Figs. 4
and 5
. Thus, one could speculate
that the P-gly and mrp1 are the major transporters of vincristine in
these and other sensitive tissues and other ABC transporters that may
be present are not involved in the export of this agent. In contrast,
transport systems other than the P-gly and mrp1 may play a greater role
in the efflux of etoposide, resulting in the much lower increase in the
hypersensitivity to the epipodophyllotoxin in
mdr1a/1b(-/-), mrp1(-/-) knockout animals
that has been observed.
It cannot be stated unequivocally from the etoposide and vincristine mortality data that death was secondary to bone marrow toxicity, as opposed to another cause, such as systemic shock related to gastrointestinal toxicity. However, the degree of apparent bone marrow compromise, as manifested by the peripheral WBC count and by histological examination, was marked. Concern over therapy-limiting bone marrow toxicity in the setting of ABC transporter inhibition has been raised by others (4) . The findings presented in the this report regarding the triple knockout genotype amplify these concerns, particularly in light of the observed enhanced hypersensitivity obtained with vincristine, which normally does not express overt toxicity to the bone marrow or the gastrointestinal mucosa. Importantly, the degree of toxicity to these tissues varied with the class of drug, further supporting the concern that attempts at inhibiting the P-gly and/or MRP1 in resistant tumors in humans need to be conducted with caution because the associated toxicities are likely to vary, not only with respect to the therapeutic agent being used, but also with respect to the inhibitor and the extent to which drug export via the P-gly and MRP1 is reduced.
|
FOOTNOTES |
|---|
1 R. A. F. is a Special Fellow of the Leukemia
and Lymphoma Society. 
2 To whom requests for reprints should be
addressed, at Department of Pharmacology, Yale University School of
Medicine, 333 Cedar Street, New Haven, CT 06520. Phone:
(203) 785-4533; Fax: (203) 737-2045; E-mail: alan.sartorelli{at}yale.edu
3 The abbreviations used are: MDR, multidrug
resistance; ABC, ATP binding cassette; P-gly, P-glycoprotein; MRP1,
multidrug resistance (-associated) protein; GSH, glutathione;
mrp1(-/-), mrp1 gene deficiency;
mdr1a/1b(-/-), mdr1a/1b gene
deficiency; VBL, vinblastine; mdr1a/1b(-/-),
mrp1(-/-), combined mdr1a/1b,
mrp1 gene deficiency; MTS,
3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4
sulfophenyl)-2H-tetrazolium; MTD, maximum tolerated
dose.
Received 8/ 9/00. Accepted 12/13/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. S. Lagas, M. L.H. Vlaming, and A. H. Schinkel Pharmacokinetic Assessment of Multiple ATP-binding Cassette Transporters: The Power of Combination Knockout Mice Mol. Interv., June 1, 2009; 9(3): 136 - 145. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Hopper-Borge, X. Xu, T. Shen, Z. Shi, Z.-S. Chen, and G. D. Kruh Human Multidrug Resistance Protein 7 (ABCC10) Is a Resistance Factor for Nucleoside Analogues and Epothilone B Cancer Res., January 1, 2009; 69(1): 178 - 184. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. B. Goralski, P. D. Acott, A. D. Fraser, D. Worth, and C. J. Sinal BRAIN CYCLOSPORIN A LEVELS ARE DETERMINED BY ONTOGENIC REGULATION OF MDR1A EXPRESSION Drug Metab. Dispos., February 1, 2006; 34(2): 288 - 295. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. G. Belinsky, P. A. Dawson, I. Shchaveleva, L. J. Bain, R. Wang, V. Ling, Z.-S. Chen, A. Grinberg, H. Westphal, A. Klein-Szanto, et al. Analysis of the In Vivo Functions of Mrp3 Mol. Pharmacol., July 1, 2005; 68(1): 160 - 168. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Mahadevan and A. F. List Targeting the multidrug resistance-1 transporter in AML: molecular regulation and therapeutic strategies Blood, October 1, 2004; 104(7): 1940 - 1951. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. De Rosa, D. Sillence, C. Ackerley, and C. Lingwood Role of Multiple Drug Resistance Protein 1 in Neutral but Not Acidic Glycosphingolipid Biosynthesis J. Biol. Chem., February 27, 2004; 279(9): 7867 - 7876. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Mochida, K.-i. Taguchi, S. Taniguchi, M. Tsuneyoshi, H. Kuwano, T. Tsuzuki, M. Kuwano, and M. Wada The role of P-glycoprotein in intestinal tumorigenesis: disruption of mdr1a suppresses polyp formation in ApcMin/+ mice Carcinogenesis, July 1, 2003; 24(7): 1219 - 1224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Burg, P. Wielinga, N. Zelcer, T. Saeki, G. J. Mulder, and P. Borst Inhibition of the Multidrug Resistance Protein 1 (MRP1) by Peptidomimetic Glutathione-Conjugate Analogs Mol. Pharmacol., November 1, 2002; 62(5): 1160 - 1166. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. P. Lin, D. R. Johnson, R. A. Finch, M. G. Belinsky, G. D. Kruh, and A. C. Sartorelli Comparative Study of the Importance of Multidrug Resistance-associated Protein 1 and P-Glycoprotein to Drug Sensitivity in Immortalized Mouse Embryonic Fibroblasts Mol. Cancer Ther., October 1, 2002; 1(12): 1105 - 1114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Zelcer, T. Saeki, G. Reid, J. H. Beijnen, and P. Borst Characterization of Drug Transport by the Human Multidrug Resistance Protein 3 (ABCC3) J. Biol. Chem., November 30, 2001; 276(49): 46400 - 46407. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
); mrp1(-/-) (
);
mdr1a/1b(-/-) (•); and
mdr1a/1b(-/-), mrp1(-/-) (
)
embryonic fibroblasts were exposed to various concentrations of
vincristine or paclitaxel for 96 h or to etoposide for 72 h.
Drug sensitivities were evaluated using the MTS/PES colorimetric assay.
Data points represent means from three independent experiments;
bars, SE.
); mdr1a/1b(-/-) (