Download Increased expression of the multidrug resistance-associated protein

From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
1995 85: 186-193
Increased expression of the multidrug resistance-associated protein
gene in relapsed acute leukemia
E Schneider, KH Cowan, H Bader, S Toomey, GN Schwartz, JE Karp, PJ Burke and SH
Kaufmann
Updated information and services can be found at:
http://www.bloodjournal.org/content/85/1/186.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
Increased Expression of the Multidrug Resistance-Associated Protein Gene
in Relapsed Acute Leukemia
By Erasmus Schneider, Kenneth H. Cowan, Helaine Bader, Sara Toomey, Gretchen N. Schwartz, Judith E. Karp,
Philip J. Burke, and Scott H. Kaufmann
Quantitative reverse transcriptase polymerase chain reaction (RT-PCR) was used to determine relative levels of
transcripts for MDRl and the recently described multidrug
resistance-associated protein (MRP) in normal lymphohematopoietic cells andin 62 bone marrow aspirates of newly
diagnosed and recurrent acute leukemia. Levels of MRP expression in newly diagnosed AML samples were similar to
those observedin normal bone marrow cells (CDBCnegative
and CDSCpositive) and in unselected HL60 human promyelocytic leukemiacells, which were used as an internal control throughout this study. In contrast, samples of AML obtained at the time ofrelapsecontainedapproximately
twofold higherlevels of MRP RNA (P c .01). Analysisof
paired samples, the first obtained at diagnosis and the second at relapse, from 13 acute myelogenous leukemia(AML)
and four acute lymphocytic leukemia (ALL)
patients showed
that MRP expression was increased at the time of relapse
in greater than 80% of patients. In contrast, no consistent
changesof MDRl expression at relapse were observed.
These results raise the possibility that increased MRP expression might contribute to leukemic relapse.
This is a US government work. There are no restrictionson
its use.
ITH CURRENT induction regimens, 60% to 80% of
patients with newly diagnosed acute myelogenous
leukemia (AML) or acute lymphocytic leukemia (ALL)
achieve a complete remission (CR). Despite this initial success, only 20% to 30% of adult patients with AML and 10%
to 20% of adult patients with ALL are cured.’” The remainder of patients either never achieve a CR or relapse after
initially achieving a remission. These refractory or relapsed
leukemias often respond poorly to chemotherapy and display
resistance to a broad spectrum of antineoplastic agents. Identifying the molecular mechanisms responsible for the failure
of chemotherapy and the emergence of multiple-drug resistance (MDR) in relapsed patients isan area of intensive
investigation.
Several mechanisms have been proposed for the development of MDR. In vitro and in vivo, the best studied mechanism involves overexpression of the MDRI gene product,
P-glycoprotein (Pgp)!-6 While it has been relatively straightforward to prove that elevated expression of Pgp causes
resistance to anthracyclines, anthracenediones, epipodophyllotoxins, vinca alkaloids, and other agents in drug-selected
tissue culture cell lines, directly proving that Pgp expression
plays a role in clinical drug resistance has been more difficult. For example, several studies have suggested that MDRl
expression is a prognostic factor in AML”” and ALL,I2
whereas other studies have suggested that Pgp is rarely expressed in AMLL3*L4
or ALL” samples, even at the time of
relapse. Furthermore, studies using modulators of Pgp have
suggested that the effect of Pgp on anthracycline accumulation is modest even in those samples that contain significant
levels of Pgp.’o.’i.‘6”8
Thus, the role of Pgp expression as a
prognostic factor and as a cause for relapse in acute leukemia
remains an area for active i n v e s t i g a t i ~ n . ~ ~ - ~ ~
It was recently suggested that overexpression of another
gene product, the multidrug resistance-associated protein
(MRP), also contributes to MDR in tissue culture cells.23The
full-length cDNA for MRP was isolated from the multidrug
resistant human small-cell lung cancer cell line H69AR. It
appears to encode an adenosine triphosphate (ATP)-binding
(drug)-transport protein that is distinct from P ~ P . ’Overex~
pression of this gene has also been detected in several other
MDR cell lines, including two independently derived Pgpnegative, doxorubicin-resistant HL60 human acute promyelocytic leukemia cell line variants (HL60IAR and HL60/
Adr).24-28
In many of these MRP-overexpressing cell lines,
drug accumulation was reduced, and/or intracellular drug
distribution was altered. Furthermore, recent studies demonstrated that transfection and expression of a cDNA for MRP
in cells increased the resistance of the transfected cells to
natural product drugs, including doxorubicin, etoposide, and
v i n c r i ~ t i n eThus,
. ~ ~ ~overexpression
~~
of MRP can lead to an
MDR phenotype in vitro. However, to date there has been
no evidence for a role for MRP expression in clinical drug
resistance.
In a recent studyi6we used a functional assay to probe
for Pgp function in bone marrow aspirates from adults with
newly diagnosed and relapsed AML. Evidence of Pgp function (defined as a 2 20% increase in nuclear daunorubicin
accumulation in the presence of the Pgp modulator quinidine) could be detected in only 10% of cases of newly diagnosed AML, and the incidence of the Pgp phenotype did not
appreciably increase at the time of relapse. These studies
have now been extended by measuring expression of MDRl
and MRP mRNA by quantitative reverse transcriptase polymerase chain reaction (RT-PCR) in paired samples obtained
from the same patients at the time of initial diagnosis and
again after relapse. The observation that MRP expression
increased in more than 80% of samples at the time of relapse
compared with diagnosis raises the possibility that MRP
might playa role in drug resistance in human acute leukemia.
From the Medicine Branch, National Cancer Institute, Bethesda;
the Oncology Center, Johns Hopkins Hospital, Baltimore; and the
Department of Pharmacology and Molecular Sciences, Johns Hopkins School of Medicine, Baltimore, MD.
Submitted July 5, 1994: accepted September I, 1994.
S.H.K. is a Leukemia Society of America Scholar.
Address reprint requests to Erasmus Schneider, PhD, Medicine
Branch, National Cancer Institute, Bldg 10, Rm 12N226, 9000 Rockville Pike, Bethesda, MD 20892.
The publication costsof this article were defrayedin part by page
chargepayment. This ariicle must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
This is a US govemment work. There are no restrictions on its use.
0006-4971/95/8501-0002$0.00/0
186
Blood, Vol 85,No 1 (January l), 1995:pp 186-193
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
107
MRP EXPRESSION IN AML AND ALL
MATERIALS AND METHODS
Cell lines and clinical samples. Wild-type HL60 cells derived
from an AML patient” were obtained from American Type Culture
Collection (ATCC; Rockville, MD). HL60/AR cells, a Pgp-negative
multidrug-resistant subline selected for Adriamycin resistance, were
a gift from S. Grant (Columbia University, New York, NY).”
SW620 colon cancer cells were provided by S. Bates [National
Cancer Institute (NCI), Bethesda, MD].”
Bone marrow aspirates were obtained before chemotherapy at the
time of initial diagnosis or at relapse. Immediately after harvest,
marrow mononuclear cells were isolated by sedimentation on FicollHypaque gradients and solubilized in guanidinium thiocyanate under
reducing conditions.I6 The samples analyzed inthe present study
were selected from a larger group of sarnples16,1“based on their
high percentages of blasts at diagnosis and relapse. Examination of
cytospin preparations indicated that samples obtained at diagnosis
contained an average of 85% blasts.
Normal bone marrows were obtained from the iliac crest of normal
donors after informed consent under a human use protocol approved
by the NCI. Bone marrow aspirates (8-10 mL) were collected in
each of two 20-mL syringes containing preservative-free heparin
(Lymphomed, Deerfield, IL) as an anticoagulant and then diluted
in Hanks’ balanced salt solution (HBSS) without CaCI, or MgClz
(GIBCO, Grand Island, NY) containing penicillin and streptomycin.
Cells with a density of ZG 1.077 g/mL were collected and washed
three times with HBSS after separation on a Ficoll-sodium diatroate
gradient (LSM; Organon Teknika Corp, Durham, NC). The low
density cells were diluted to 2 X lo6cells per milliliter in an enriched
Iscove’s Modified Dulbecco’s medium (IMDM)34containing 10%
heat inactivated fetal bovine serum (Hyclone, Logan, UT) and were
incubated in tissue culture flasks overnight in a humidified37°C
incubator with 5% COz. The nonadherent cells were collected and
washed twice with HBSS. CD34+ cells were isolated by positive
immunomagnetic selection using a high-gradient magnetic separation column and MiniMacs CD34 progenitor cell isolation kit (Miltenyi Biotec Inc. Sunnyvale, CA). The purity of the recovered cells
was determined by staining with CD34 (anti-HPCA-2) fluorescein
isothiocyanate (FITC; Becton Dickinson, San Jose, CA), which binds
to a distinct epitope on the CD34 molecule. In the present studies,
flow cytometry analysis showed that the purity of CD34+ cells in
the CD34+ cell fractions was 90% to 98% and less than 1% in the
CD34- fractions.
Normal peripheral blood cells obtained from a healthy donor by
Ficoll gradient separation were separated into lymphocyte subfractions as previously described.”
Adult patients with AML (median age, 43 years; range, 18 to 75
years) were subsequently treated on protocol JH8410 (cytosine arabinoside (Ara-C)/daunorubicin/4’“[9-acridinylamino]methanesulfon-m-aniside [ I ~ A M S A ] )or~ ~JH8914 [granulocyte-macrophage colony-stimulating factor (GM-CSF)/Ara-C/daunorubicin/
etoposide] as recently described.I6Adult patients with ALL (median
age, 30 years; range, 18 to 74 years) were treated with prednisone,
vincristine, etoposide, and L-asparaginase followed by Ara-C and
daunorubicin on protocol JH8802 as recently described.’& Patients
with fewer than 5% blasts in the marrow and recovery of normal
marrow function that persisted at least until the time of consolidation
therapy were considered to have achieved CRs. Patients who died
of toxic complications (infection or bleeding) before the time of
expected marrow recovery were considered not eligible for evaluation. All other patients were considered nonresponsive (NR).
Quantitative RT-PCR. Total RNA was prepared according to
the method of Chomczynski and Sac~hi.~’
RT-PCR was performed
as previously d e s ~ r i b e d .Total
~ ~ . ~RNA
~
(1 pg) from each AML or
ALL sample or each cell line was reverse transcribed in 20 pL of
RT buffer [ I O mmol/L TRIS-HCI, pH 8.3; 50 mmol/L KCI; 5 mmol/
L MgClz; 1 mmoVL each of deoxyadenosine triphosphate (dATP),
deoxyguanidine triphosphate (dGTP), deoxycytidine triphosphate
(dCTP), and (deoxythymidine triphosphate ( d m ) ] containing 2 U/
pL RNase inhibitor, 0.003 AZmunits random hexanucleotides (both
from Boehringer Mannheim, Indianapolis, IN), and 0.4 UIpL AMV
reverse transcriptase (Promega, Madison, WI). The reaction conditions were 25°C for 10 minutes, 42°C for 15 minutes, and 99°C for
5 minutes. The resulting cDNA mixture was diluted twofold, fivefold, and 50-fold (or twofold, 10-fold, and 100-fold) in RT buffer.
Target sequences for MDRI, MRP, and glyceraldehyde-3-phosphate
dehydrogenase (G3PDH) were separately amplified in the twofold,
fivefold, and 50-fold dilutions, respectively, for 25 cycles (95°C for
10 seconds and 60°Cfor 15 seconds in a 9600 Thermocycler [PerkinElmer Cetus, Norwalk, CT]) by the PCR with specific primers for
each gene, using the hot start m~dification.~’Digoxigenin-ll-2‘deoxyuridine-5’-triphosphate(0.1 pL; Boehringer Mannheim) was
added to each tube of PCR mixture containing 12.5 mmoVL TRIS,
pH 8.3; 62.5 mmol/L KCI; 2.5 mmol/L MgClz; 0.2 mmoVL each
of dATP, dGTP, dCTP, and dlTP; 2.5 U/pL Taq polymerase (Boehringer Mannheim); and 0.3 pmoVL of each specific primer pair in
a total volume of 100 pL. An aliquot (15 pL) of the PCR products
was separated on a 2% agarose gel in TRIS-borate-EDTA buffer. The
gels were denatured and neutralized, and the DNA was transferred in
10 x SSC (1.5 moUL sodium chloride, 150 mmol/L sodium citrate
solution) onto a nylon membrane (Boehringer Mannheim). The
membranes were then processed for chemilumenescence detection
of the PCR products by the Genius system according to the manufacturer’s instructions (Boehringer Mannheim). The relative amounts
of each PCR product for MRP and MDRI were quantitated by densitometry and were normalized relative to the amount of PCR product
for G3PDH that had been amplified in parallel reactions. The following primers wereused (with expected size of the PCR product):
MRP1, nucleotide (nt) 792 to 816; MRP2, nt 1062 to 1086 (285
bp); MRP3, nt 3652 to 3676; MRP4, nt 4160 to 4184 (533 bp)23;
MDRI-1, nt 410 to 441; MDRI-2, nt 664 to 695 (286 bp)38;G3PDH1, nt 75 to 100; G3PDH-2, nt 670 to 696 (621 bp).@These primers
were purchased from The Midland Reagent Company (Midland,
TX). For MRP, either primer pair MRPI/MRP2 or MRP3/MRP4
was used with identical results. The specificity of the MRP primer
pairs was confirmed by direct sequencing of the resulting PCR products, using the Circumvent sequencing kit from New England Biolabs (Beverly, MA).
Southern blotting. Total genomic DNA from paired leukemia
samples was isolated by SDS lysis and proteinase K digestion, followed by phenokhloroform extraction and ethanol precipitation.
Purified DNA (10 pg) was digested with EcoRI overnight and electrophoresed on a 0.7% agarose gel. DNA from HL60 and HL6O/AR
cells was treated identically and provided control samples of an
unamplifiedand an amplified MRP gene on the same blot. The
DNA was denatured, transferred onto a nylon membrane (Boehringer
Mannheim) and hybridized at 65°C with a PCR-generated digoxigenin-labeled DNA probe consisting of the first 890 bp of the coding
region of the MRP cDNA. After hybridization the membrane was
washed in 0.5 X SSC/O.I% sodium dodecyl sulfate (SDS) at 65”C,
and bound probe was detected by the Genius chemilumenescence
method (Boehringer Mannheim) according to manufacturer’s instructions.
RESULTS
Validation of PCR assay. The yield of PCR product is
proportional to the starting amount of the template only under conditions in which PCR amplification proceeds exponentially. To establish conditions under which the amount
of PCR product wouldbe directly proportional to the amount
of template used, cDNA was synthesized from total
RNA
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
1aa
SCHNEIDER ET AL
A
Dilution
25 cycles
27 cycles
v!
0 0
v!
0 0
oowtu
0 0
N m r N
N m r N
MRP
Dilution
0 0
oowtu
*corm
*corm
GSPDH
.-
C
; 10000
.*,
L
f!
m
Y
U
C
m
I B). The observation that the curves were linear and parallel
with a slope of approximately I indicated that this method
could be used to compare levels of MRP expression between
different samples. Furthermore, these studies also demonstrated that a fivefold to IO-fold dilution for MRP (corresponding to 0.1 to 0.2 pg RNA) and a SO-fold to 100-fold
dilution for G3PDH (corresponding to 0.01 to 0.02 pg RNA)
gave amplified signals whose intensity was within the linear
range of the reaction. These dilutions wereused for the
analysis of MRP expression in all of the samples described
below.
MRP expression in normal cells. To provide a baseline
for comparison, MRP expression was analyzed by quantitative RT-PCR in normal B cells, T cells, and granulocytes as
well as in normal CD34- andCD34'bonemarrow
cells.
Results were normalized to levels of expression of the
housekeeping gene G3PDH that had beendetermined in parallel. HL60 human promyelocytic leukemia cells were included in each experiment to provide an internal standard.
MRP mRNA levels in HL60 cells were found to be similar
to the average MRP mRNA level observed in 55 different
unselected human tumor cell lines (A. Fojo, personal communication, May 1993). The MRP/G3PDH mRNA level in
HL60 cells was arbitrarily assigned a value of 1 .O, and levels
of all other samples were compared with this value. All
normal lymphohematopoeitic cell populations analyzed had
levels of MRP mRNA that were similar to those in HL60
cells, with the exception of T cells, which had twofold higher
MRP mRNA levels (Fig 2).
MRP expression in A M L and ALL. Using the same approach, expression of MRP mRNA wasanalyzed in 43 AML
(29 collected at initial diagnosis and 14 at time of disease
m
0.001
0.01
0.1
1
NI RNA
Fig 1. Validation of PCR assay. (A) From HL60 cells 1 p g of total
RNA was reverse transcribed. The resulting cDNA mixture was serially twofold diluted from 2.5- t o 320-fold. Dilutions 2.5 t o 20 were
then amplified with MRP primers 1 and 2 and dilutions 40 t o 320
with G3PDH primers for25 or 27 cycles. Of each PCR reaction, 15 p L
were analyzed by agarose gel electrophoresis and transferred onto
a nylon membrane, and the products weredetected by chemilumenescence as described in Materials and Methods.(B) Each band was
then quantitated by densitometry, and the results were graphed
against the calculated amount of RNA present in the originalsample.
m, MRP 25 cycles, slope = 1.07, r = 0.996; A, MRP 27 cycles, slope =
1.14, r = 0.994; 0, G3PDH 25 cycles,slope = 1.13, r = 0.999; V, GJPDH
27 cycles, slope = 1.04, r = 1.000.
V.V
4
m
from HL60 cells by reverse transcription. The cDNA was
then serially diluted, and each dilution was separately amplified using specific primer pairs for MRP or G3PDH. The
resulting products were separated by gel electrophoresis and
transferred onto a nylon membrane. After transfer, the membranes were processed for chemilumenescence detection of
the PCR products, which were quantitated by densitometry
of the bands after exposure to x-rayfilm (Fig 1A). The
relative band intensities were then plotted against the calculated amount of RNA originally present in each dilution (Fig
n
0
+
d
m
n
0
v)
Q,
0
m
-
v)
Q,
V
I-
Fig 2. Expression of MRP in normal bone marrow and peripheral
blood cells. MRP mRNA levels were determined by RT-PCR as described in Materials and Methods, were normalized for
G3PDH mRNA
levels, and are expressed relative t o the level of MRP mRNA in HL60
cells, which was arbitrarily set at 1.0. For comparison, the relative
level of MRP mRNA in drug-resistant HL6O/AR cells was 9.5.
N
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
MRP 189
IN AML AND ALL
Table 1. MRP and MDRl Expression in AML and ALL Samples
MDRl Expressiont
MRP Expression'
All samples
At diagnosis
At relapse
P value*
Paired samples only
At diagnosis
At relapse
P value*
AML
ALL
AML
ALL
1.0 2 0.1 (N = 29)
1.7 5 0.3 (N = 14)
P .01
1.5 2 0.5 (N = 15)
1.4 t 0.1 (N = 4)
NS
1.0 i 0.3 (N = 17)
0.9 i 0.2 (N = 9)
NS
1.2 2 0.7 (N = 5 )
1.0 t 0.6 (N = 4)
NS
1.0 2 0.2 (N = 13)
1.8 2 0.3 (N = 13)
P < .01
0.7 2 0.2 (N = 4)
1.4 2 0.1 (N = 4)
P < .05
1.7 2 0.6 (N = 6)
0.8 2 0.3 (N = 6)
NS
1.9 2 1.0 (N = 3)
1.2 2 0.9 (N = 3)
NS
MRP and MDRl expression was determined by quantitative RT-PCR as described in Materials and Methods and normalized with G3PDH
expression determined in parallel reactions. Each individual sample was analyzed at least twice, and the average of each sample was used to
calculate the expression relative to that in HLW or SW620 cells. These relative expression levels were used to calculate the means (ItSE)
presented in this table. For comparison, MRP expression in HL60 cells was 1.0; in HLGO/AR cells, 9.5; and in SW620 cells, 1.8. MDRl expression
in HL60 and HL6O/AR cells was undetectable and in SW620 cells was 1.0.
Abbreviation: NS, not significant.
Relative to MRP expression in HL60 cells.
t Relative to MDRl expression in SW620 cells; only samples with quantifiable amounts of MDRl were used to calculate means.
Student's t-test.
*
relapse) and 19 ALL (15 collected at diagnosis and 4 at
relapse) samples. Results were again normalized to levels of
expression of the housekeeping gene G3PDH that had been
determined in parallel and were compared with HL60 cells
(Table 1). M W expression in the clinical samples ranged
from undetectable to 8.6-fold greater than in HL60 cells,
with a mean relative expression level of 1.O at diagnosis and
1.7 at relapse in AML (median, 0.9 and 1.5; range, 0.01 to
3.1 and 0.4 to 5.0,respectively; P < .01)and 1.5 at diagnosis
and 1.4 at relapse in ALL (median, 0.8 and 1.4; range, 0.4
to 8.6 and 1.2 to 1.5, respectively). It should be noted that
M W expression levels in ALL samples at diagnosis were
distributed over a wide range, from as low as 0.4 to as high
as 8.6, and that corresponding samples at relapse from the
two highest expressors were not available. This resulted in
artificially distorted mean values. By comparison, in the doxorubicin-selected, Pgp-negative, multidrug-resistant HL601
AR variant that is more than 100-fold resistant to doxorubicin, M W expression was 9.5-fold higher than in the parental
HL60 cell line.
There was no clear relationship between M W expression
at diagnosis and leukemia type (Fig 3A), response to therapy
(Fig 3B), and duration of remission (Fig 3C). Furthermore,
there was also no relationship between M W expression and
expression of the early myeloid marker CD34 on the leukemia cells (median M W expression was 1.1 if CD34- and
1.0 if CD34+), in agreement with our observation that M W
expression in CD34- and CD34+ normal marrow cells was
similar. In contrast, a statistically significant ( P < .01) increase in M W expression was observed when samples from
C
B
0
0
:3t0
Y
LT
I
IGi
0
0
1
2
3
4
v
5
FAB
n
6
7
0
8
E
T
cell cell
CR
NR
InitialResponse
to
NR
CR
Rx
200
400
Remission
Duration
600
800
+l000
(d)
Fig 3. Expression of MRP mRNA in newly diagnosed leukemia. MRP mRNA levels were determined by RT-PCR as described in Materials
and Methods, were normalized for GBPDH mRNA levels, and are expressed relative to the level of #RP mRNA in HL80 cells, which was
arbitrarily set at 1.0. Relative MRP mRNA expression was examined as a function of French-American-British(FA61 classlfkation (Al. mponse
, OO days indicae actual duration of remission. For comperison,
to initial thorapy (B), or duration of remission (C). All time intervals IN. than 1O
the relative level of MRP mRNA in drug-resistant HLGO/AR cells was 9.5. (C) 0, AML; 0, ALL. NR, no response; DX, diagnosis; Rx, therapy.
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
190
SCHNEIDER ET AL
B
A
P
DX
4 r
Relapse
ox
C
MRP
4
."
G3PDH
, -
-
1_
relapsed patients were compared with samples from patients
with newly diagnosed leukemia (Table 1). From a total of
29 AML and 15 ALL patients with newly diagnosed acute
leukemia, paired samples (ie, the first harvested at the time
of diagnosis and the second at the time of leukemic relapse)
were available from 17 ( l 3 AML and four ALL).When
MRP expression at diagnosis and recurrence was compared
in each pair separately (Fig 4A), the mean increase in MRP
expression at relapse was 2.8-fold in AML and 2.3-fold in
ALL (median, 1.5- and2.2-fold, respectively). Of the 17
paired samples, one (6%) showed a decrease at relapse, and
two (12%) remained equal (less than 220% change). However, the vast majority of paired samples (82%) showed an
increase in MRP expression at the time of relapse. In those,
expression at relapse ranged from 1.2- to 15.3-fold that at
the time of diagnosis (one paired sample was not quantifiable
due to a lack of detection at diagnosis). In summary, MRP
was detected in all but one of the leukemia samples analyzed,
and there appeared to be a tendency for higher MRP expression at the time of relapse.
MDRl expression in AML and ALL. Expression of
MDRl mRNA was also analyzed by quantitative RT-PCR
in all 43 AML and 19 ALL samples. Expression levels were
normalized to those of the housekeeping gene G3PDH and
then compared with M D R l expression in SW620 cells, a
cell line that naturally expresses low levels of MDR133(Table
1). No MDRI expression was detected in HL60 and HL60/
Relapse
Fig 4. Expression of MRP (A, C) and MDRT (6)
mRNA in paired samples obtained at diagnosis and
after the first relapse. MRP and MDRT mRNA levels
were determined by quantitative RT-PCRasdescribed in Materials and Methods, were normalized
for GBPDH mRNA levels, and are expressed relative
to the level of MRP mRNA in HL60 cells or MDRT
levels in SW620 cells. MRP mRNA was quantifiable
in all '17 pairs, whereas MDRT mRNA was quantifiable in nine paired samples and was too low to be
accurately quantitated in one or both partners of
seven pairs (data not shown). 0, AML; 0, ALL; horizontal bars; mean relative expression levels with
standard errors. Mean levels for MRP were 0.95 2
0.1 at diagnosis and 1.7 t 0.25 at relapse, P = .0015;
for MDRT, mean levels were 1.7 ? 0.5 and 0.9 2 0.3,
respectively, P > .l (C) Film exposure of PCR products for MRP in selected paired samples. G3PDHwas
used for normalization and is shown for comparison.
Pairs 1 to 3 are AML, pair 4 is ALL, and last pair is
sensitive HL60 and resistant HLGO/AR cells.
AR cells. In contrast, in the KGla human leukemia cell
line that had previously been shown to exhibit a twofold to
threefold decrease in LDS,, for daunorubicin and etoposide
in the presence of Pgp modulator^,'".^' MDRI expression
was 2.8-fold higher than in SW620 cells.
In 27 of the 62 clinical samples (17 of 43 AML and 10
of 19 ALL), M D R l expression was too low to bequantitated.
In the remaining 35 clinical samples, relative expression of
M D R l ranged from 0.05-fold to 3.7-fold the level expressed
in SW620 cells, with a mean relative expression level of 1.1
at diagnosis and 0.9 at relapse in AML (median, 0.6 and
0.8; range, 0.05 to 3.7 and 0.2 to 2.0, respectively) and 1.2
at diagnosis and 1.O at relapse in ALL (median, 0.9 and 0.5;
range, 0.1 to 3.7and0.3 to 2.9, respectively). Consistent
with our previous results obtained using a functional assay
for Pgp," only six of 62 samples manifested a level of M D R l
expression that approached the level observed in KGla cells.
There was no correlation between MDRI expression and
leukemia type (Fig 5A), response to therapy (Fig 5B), or
duration of remission (data not shown). Furthermore, there
was also no correlation between expression of MDRI and
MRP at diagnosis (r = 0.3, P = .2). In contrast to the results
withMRP, there was also no clear pattern of change in
MDRI levels at the time of relapse (Fig 4B). Of the 17
paired samples, nine (six AML and three ALL) were informative, with a mean increase at the time of relapse of 1.1fold in AML and 0.8-fold in ALL (median, 0.6- and0.8-
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
MRPEXPRESSION IN AML ANDALL
191
ALL
0
Q
CR
NR
Response t o Initio1 R x
FAB
Fig 5. Expression of MDRl mRNA in newly diagnosed leukemia.
Relative MDRl expression was examined as a function of FrenchAmerican-British classification (A) or response t o initial therapy (B).
MDR7 mRNA levels were determined by quantitativeRT-PCR as described in Materials andMethods, were normalized for
GBPDH mRNA
levels, and are expressed relative to the level of MDRl mRNA in
SW620 cells. For comparison, MDRl was undetectable in HL60 and
HL6O/AR cells,and expression levels in KGla cells were 2.8-fold those
in SW620 cells.KGla cells show a 60% increase in daunorubicin accumulation in the presence of Pgp modulators. NR, no response; DX,
diagnosis; Rx, therapy.
fold, respectively). M D R l expression at relapse ranged from
0.05- to 3.5-fold that at the time of diagnosis, with one
sample ( 1 1%) remaining unchanged (less than 520%
change), three (33%) increased, and five (55%) decreased.
In conclusion, M D R l expression did not appear to correlate
with any of the clinical parameters analyzed. Although samples with MDRI mRNA levels too low to be quantitated by
densitometry were excluded from this analysis, the results
remained qualitatively similar when relative MDRI expression in these samples was arbitrarily set at 0.01 (1% of
SW620). Thus, we do not believe that the omission of samples withverylow
M D R l levels has led to an artificial
distortion of the results.
Genomic DNA analysis. In several drug-selected cell
lines that overexpress MRP, amplification of the MRP gene
has been described (Schneider et al. unpublished data, June
I 993).2.?..’S.ZRTherefore, we analyzed genomic DNA from five
paired samples (three AML and two ALL) in which MRP
mRNA levels were 1.2- to 2. l-fold higher at relapse than
at diagnosis (in one pair, no MRP mRNA was detected at
diagnosis, but was clearly detectable at relapse; Fig 6). For
comparison, we also analyzed genomic DNA from HL60
cells and their drug-resistant derivative HL6O/AR cells,
which had 9.5-fold higher MRP mRNA levels. While there
was clear evidence for amplification of the MRP gene in
HL60/AR cells, we did not find any indication that the MRP
gene is amplified in the relapsed leukemia samples.
DISCUSSION
In a previous study of marrow samples from patients with
newly diagnosed and relapsed AML, we reported that drug
sensitivity did not correlate with Pgp function or topoisomerase I1 levels.’6 In this report these earlier studies were extended by analyzing levels of MDRI and MRP mRNA by
quantitative RT-PCR. In agreement with our previous results, there was no correlation between M D R l expression
and response to therapy. Moreover, comparison of samples
obtained at the time of diagnosis and again at the time of
relapse did not show any consistent changes in M D R l expression. In addition, only six of 62 samples (10%)exhibited
M D R l mRNA levels that had been shown to be functional
in the KGla cell line, whereas M D R l mRNA levels were
below levels of quantification in 45% of the samples. Thus,
our results tend to suggest that Pgp expression had little or
no impact on the clinical outcome in this cohort of leukemia
patients.
In contrast, MRP mRNA could be readily detected and
quantitated in all but one leukemic sample. Moreover, there
was a statistically significant (P < . O l ) increase in the average MRP expression level at the time of relapse compared
with that at the time of diagnosis (Table 1). This result is in
agreement with observations by Hart et a14’ andZhouet
al;’ who also found a slight (albeit statistically insignificant)
increase in MRP expression in relapsed compared with
newly diagnosed AML. Furthermore, increased MRP expression at the time of leukemic recurrence was also observed
in 14 of 17 cases when we directly compared paired samples
(Fig 4A). Therefore, our results are consistent with the hypothesis that increased MRP expression might contribute to
leukemic relapse.
Despite the observation that MRP mRNA levels tended
to increase at the time of relapse, there wasno obvious
correlation between pretreatment MRP mRNA levels and
response to therapy or duration of remission. This lack of
correlation between pretreatment mRNA levels and clinical
kb
-23.1
- 9.4
- 6.6
- 4.4
- 2.3
- 2.0
Fig 6. Southern blot analysis of genomic DNA from paired Samples obtained at diagnosis and after the first
relapse. Total genomic
DNA was isolated from selected paired samples and digested with
€COW. After electrophoresis and transfer onto a nylon membrane,
the DNA was hybridized with a 890-bp fragment from the 5‘-end of
the coding regionof MRP. Detection of the hybridized probe was by
chemilumenescence. For comparison, DNA from HL60 and HL6O/AR
cells was also analyzed. The samples are arranged in pairs, with the
left lane of each pair obtainedat diagnosis and the right
lane obtained
at relapse. Pairs 291/TPT5 and 39/49 are ALL; pairs 132/139,79/144,
and 358/TPT12 are AML. Equal loading was confirmed by ethidium
bromide staining.
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
192
response could result from variations in one or more factors
that might differ among different leukemic marrows, notably
posttranscriptional regulation of protein expression, protein
stability, and cell-to-cell heterogeneity of gene expression.
Further studies at the single cell level are required to address
some of these issues.
The question arises whether the moderate increases in
MRP expression at relapse are sufficient to confer drug resistance. Studies in drug-selected cell lines indicate that 10- to
15-fold increases in MRP mRNA levels can result in up to
100- to 200-fold resistance to doxorubicinz6and 30- to 150fold resistance to e t ~ p o s i d e . ~ In
~ . ' addition,
~
Grant et alz9
reported that a 50% increase in MRP RNA levels in cells
transfected with an M W expression vector resulted in a 50%
inhibitory concentration that was50% higher for doxorubicin
despite the observation that protein synthesis from the
transfected RNA appeared less efficient thanfrom the endogenous RNA. Therefore, even the moderate increases in MRP
mRNA levels observed in the relapsed leukemia samples
might be sufficientto convey increased drug resistance.
When one considers that antileukemia agents are administered at maximallytolerated doses, even this modest increase
in resistance might result in diminished clinical responsiveness.
The present studies suggest that M W gene amplification
does notplay a role in the increased MRP expression in
clinical samples (Fig 6). Instead, other mechanisms need to
be considered. For instance, it is possible that MRP was
heterogeneously expressed in the original leukemic cell population and that MRP-overexpressing subclones have selectively survived the chemotherapeutic regimens used in the
present study. Alternatively, it is possible that MRP expression has been directly induced by the drugs used during
therapy, in much the same waythat MDRl appears to be
induced in certain celllines." Further studies on the frequency of MRP-positive cells and on the regulation of MRP
gene expression are required to distinguish between these
possibilities.
At present it is unknown whether our observations are
widely representative of MRP expression in acute leukemia.
The practical requirement that the samples in the present
study contain greater than lo7 cells with a majority of blast
cells might have resulted in a skewed population that contained a relative paucity of certain leukemic phenotypes,
such as secondary leukemias and leukemias arising in the
setting ofan antecedent myelodysplastic syndrome. Likewise, the emphasis on paired samples might have resulted
in a study that was weighted toward patients who had poor
disease-free survival. Consistent with this possibility, we
note that seven of the 17 patients (41%) providing paired
samples never achieved a CR and that the median diseasefree survival for the remaining patients providing paired
samples was only 254 days. In contrast, mean duration of
disease-free survival among the total patient population that
had achieved CR (77%) was 420 days. Finally, the chemotherapeutic regimens used in the present study might have
somehow selected for cells that preferentially expressed
MRP rather than MDRI, whereas other regimens might select for cells that express MDRI. Nevertheless, further studies of MRP expression in the context of various antileukemic
regimens appear warranted.
SCHNEIDER ET AL
ACKNOWLEDGMENT
We thank Dr Jeffrey Moscow for critically reading the manuscript
andDr Fran Hakim for FACS analysis of the CD34 cells. The
technical assistance of Sharon McLaughlin, Sandra Kiesewetter, Tim
Soos, Lisa Prichard, and Janet Szabo are gratefully acknowledged.
This study was made possible by the skillful care provided by Ken
Hall, Louanne Morrell, andthe attending physicians, medical fellows, house officers, and nurses of the Adult Leukemia Service of
the Johns Hopkins Hospital.
REFERENCES
1. Wiemik PH: Acute leukemias, in DeVita VT, Hellman S, Ro-
senberg SA (eds): Cancer: Principles and Practice of Oncology.
Philadelphia, PA, Lippincott, 1989, p 1809
2. Schiffer CA: Acute lymphocytic leukemia in adults, in Holland
DL, Weichselbaum RR
JF, Frei E, Bast RC, KufeDW,Morton
(eds): Cancer Medicine. Philadelphia, PA, Lea & Febiger, 1993, p
1946
3. Schiffer CA: Acute myeloid leukemia in adults, in Holland JF,
Frei E, Bast RC, Kufe DW, Morton DL, Weichselbaum RR (eds):
Cancer Medicine. Philadelphia, PA, Lea & Febiger, 1993, p 1907
4. Endicott JA, Ling V: The biochemistry of P-glycoproteinmediated multidrug resistance. Annu Rev Biochem 58:137, 1989
5 . Gottesman MM, Pastan I: Biochemistry of multidrug resistance
mediated by the multidrug transporter. Annu Rev Biochem 62:385,
1993
6. Gottesman MM: How cancer cells evade chemotherapy: Sixteenth Richard and Hinda Rosenthal Foundation Award lecture. Cancer Res 53:747, 1993
7. Sato H, Preisler H,Day R, Raza A, Larson R, Browman G ,
Goldberg J, Vogler R, Grunwald H, Gottlieb A, Bennet J, Gottesman
MM, Pastan I: MDRl transcript levels as an indication of resistant
disease in acute myelogenous leukemia. Br J Haematol75:340, 1990
8. Marie J-P, Zittoun R, Sikic BI: Multidrug resistance (mdrl)
gene expression in adult acute leukemias: Correlation with treatment
outcome and in vitro drug sensitivity. Blood 78586, 1991
9. Pirker R, Wallner J, Geissler K, Linkesch W, Haas OA, Bettelheim P, Hopfner M, Schemer R, Valent P, Havelec L, Ludwig H,
Lechner K: MDRl gene expression and treatment outcome in acute
myeloid leukemia. J Natl Cancer lnst 83:708, 1991
10. Compos L, Guyotat D, Archimbaud E, Calmard-Oriol P,
Tsuruo T, Troncy J, Treille D, Fiere D: Clinical significance of
multidrug resistance P-glycoprotein expression on acute nonlymphoblastic leukemia cells at diagnosis. Blood 79:473, 1992
11. teBoekhorst PAW, deLeeuw K, Schoester M, Wittebol S,
Nooter K, Hagemeijer A, Lowenberg B, Sonneveld P: Predominance
of functional multidrug resistance (MDR-I) phenotype in CD34+
acute myeloid leukemia cells. Blood 82:3157, 1993
12. Goasguen JE, Dossot J-M, Fardel 0, LeMee F, LeGall E,
Leblay R, Reprise PY, Chaperon J, Fauchet R: Expression of the
multidrug resistance-associated P-glycoprotein (P-l 70) in 59 cases
of de novo acute lymphoblastic leukemia: Prognostic implications.
Blood 81:2394, 1993
13. Ito Y, Tanimoto M, Kumazawa T, Okumura M, Mirishima
Y, Ohno R, Saito H: Increased P-glycoprotein expression and multidrug-resistant gene (mdrl) amplification are infrequently found in
fresh acute leukemia cells. Cancer 63:1534, 1989
14. Kato S, Ideguchi H, Muta K, Nishimura J, Nawata H: Absence
of correlation between cytotoxicity and drug transport by P-glycoprotein in leukemic cells. Eur J Haematol 47:146, 1991
15. Rothenberg ML,MickleyLA, Cole DE, Balis FM, Tsuruo
T, Poplack DG, Fojo AT: Expression of the MDRlP170 gene in
patients with acute lymphoblastic leukemia. Blood 74:1388, 1989
16. Kaufmann SH, Karp JE, Jones RJ, Miller CB, Schneider E,
From www.bloodjournal.org by guest on October 28, 2014. For personal use only.
MRP EXPRESSION IN AML AND ALL
Cowan K, Wendell K, Burke PJ: Topoisomerase 11 levels and drug
sensitivity in adult myelogenous leukemia. Blood 83:517, 1994
16a. Kaufmann SH, Karp JE, Burke PJ, Gore SD: Role of etoposide in initial therapy of adult ALL: The effect of P-glycoprotein
expression and diminished topoisomerase I1 levels. (submitted)
17. Kessel D, Wilberding C, Karanes C, Ratanatharathorn V:
Daunorubicin accumulation by human myeloblasts varying in anthracycline resistance. Cancer Lett 24:305, 1984
18. Ludescher C, Hilbe W, Eisterer W, Thaler J, Gotwald M,
) de
Hofman J: Low incidence of activity of P-glycoprotein ( ~ 1 7 0 in
novo acute lymphoblastic leukemia determined by a flow cytometric
assay. Blood 81:3505, 1993
19. Moscow JA, Schneider E, Cowan KH: Multidrug resistance,
in Pinedo HM, Longo DL, Chabner BA (eds): Cancer Chemotherapy
and Biological Response Modifiers Annual 14. Amsterdam, The
Netherlands, Elsevier Science Publishers B.V., 1993, p 98
20. Sikic BI: Modulation of multidrug resistance: At the threshold. J Clin Oncol 11:1629, 1993
21. Lum BL, Gosland MP, Kaubisch S , Sikic BI: Molecular targets in oncology: Implications of the multidrug resistance gene.
Pharmacotherapy 13:88, 1993
22. Arceci R: Clinical significance of P-glycoprotein in multidrug
resistance malignancies. Blood 81 :2215, 1993
23. Cole SPC, Bhardwaj G, Gerlach JH, Mackie JE, Grant CE,
Almquist KC, Stewart A J , Kurz EU, Duncan AMV, Deeley RG:
Overexpression of a transporter gene in a multidrug-resistant human
lung cancer cell line. Science 258: 1650, 1992
24. Krishnamachary N, Center MS: The MRP gene associated
with a non-P-glycoprotein multidrug resistance encodes a 190-kDa
membrane bound glycoprotein. Cancer Res 53:3658, 1993
25. Schneider E, Horton JK, Yang C-H, Nakagawa M, Cowan
KH: Multidrug resistance-associated protein (MRP) gene overexpression and reduced drug sensitivity of topoisomerase I1 in a human
breast carcinoma MCW cell line selected for etoposide resistance.
Cancer Res 54:152, 1994
26. Slovak ML, Ho JP, Bhardwaj G, Kurz EU, Deeley RG, Cole
SPC: Localization of a novel multidrug resistance associated gene
inthe HT1080DR4 and H69AR human tumor cell lines. Cancer
Res 53:3221, 1993
27. Zaman GIR, Versantvoort CHM, Smit JJM, Eijdems WHM,
deHaas M, Smith AJ, Broxterman HJ, Mulder NH, deVries EGE,
Baas F, Borst P: Analysis of the expression of MRP, the gene for
a new putative transmembrane drug transporter, in human multidrug
resistant lung cancer cell lines. Cancer Res 53:1747, 1993
28. Barrand MA, Heppell-Parton AC, Wright KA, Rabbitts PH,
Twentyman PR: A 190-kilodalton protein overexpressed in non-Pglycoprotein-containing multidrug-resistant cells and its relationship to the MRP gene. J Natl Cancer Inst 86:110, 1994
29. Grant CE, Valdimarsson G, Hipfner DR, Almquist KC, Cole
SPC, Deeley RG: Overexpression of multidrug resistance-associated
protein (MRP) increases resistance to natural product drugs. Cancer
Res 54:357, 1994
193
30. Kruh GD, Chan A, Myers K, Gaughan K, Miki T, Aaronson
SA: Expression complementary DNA library transfer establishes
mrp as a multidrug resistance gene. Cancer Res 54:1649, 1994
31. Gallagher R, Collins S, Trujiilo J, McCredie K, Ahearn M,
Toai S, Metzgar R, Oulakh G, Ting R, Ruscetti F, Gallo R Characterization of the continuous differentiating myeloid cell line (HL60) from a patient with acute promyelocytic leukemia. Blood 54:713,
1979
32. Bhalla K, Hindenburg A, Taub R N , Grant S: Isolation and
characterization of an anthracycline-resistant human leukemic cell
line. Cancer Res 45:3657, 1985
33. Lai G-M, Chen Y-N, Mickley La, Fojo AT, Bates S E Pglycoprotein expression and schedule dependence of adriamycin CYtotoxicity in human colon carcinoma cell lines. Int J Cancer 49:696,
1991
34. Schwartz GN: Use of gel-well culture chambers as a liquid
culture system to measure responses of hematopoietic colony-forming cells to growth factors. Int J Cell Cloning 7:360, 1989
35. Kaufmann SH, McLaughlin SJ, Kastan MB, Liu LF, Karp JE,
Burke PJ: Topoisomerase I1 levels during granulocytic maturation in
vitro and in vivo. Cancer Res 51:3534, 1991
36. Geller RB, Burke PJ, Karp JE, Humphrey RL, Braine HG,
Tucker RW, Fox MG, Zahurak M, Morrell L, Hall KL, Piantadosi
S : A two-step timed sequential treatment for acute myelocytic leukemia. Blood 74:1499, 1989
37. Chornczynslu P, Sacchi N: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.
Anal Biochem 162:156, 1987
38. Murphy LD, Herzog CE, Rudick JB, Fojo AT, Bates S E Use
of the polymerase chain reaction in the quantitation of mdr-l gene
expression. Biochem 29:10351, 1990
39. Chou Q, Russel M, Birch DE, Raymond J, Bloch W: Prevention of pre-PCR mis-priming and primer dimerization improves low
copy number amplifications. Nucleic Acids Res 20: 1717, 1992
40. Arcari P, Martinelli R, Salvatore F: The complete sequence
of a full length cDNA for human liver glyceraldehyde-3-phosphate
dehydrogenase: Evidence for multiple mRNA species. Nucleic Acids
Res 12:9179, 1984
41. Hendricks CB, Rowinsky EK, Grochow LB. Donehower RC,
Kaufmann SH: Effect of P-glycoprotein expression on the accumulation and cytotoxicity of topotecan (SK&F 104864), a new camptothecin analogue. Cancer Res 52:2268, 1992
42. Hart SM, Ganeshaguru K, Hoffbrand AV, Metha A B : Expression of the multidrug resistance-associated protein (MRP) gene in
acute myeloid leukemia (AML). Blood 82:258, 1993 (suppl, abstr)
43. Zhou D-C, Hoang-Ngoc L, Delmer A, Faussat A-M, Russo
D, Zittoun R, Marie J-P: Expression of resistance genes in acute
leukemia. Leuk Lymph 13:27, 1994 (suppl 1)
44. Chaudhary PM, Roninson IB: Induction of multidrug resistance in human cells by transient exposure to different chemotherapeutic drugs. J Natl Cancer Inst 85:632, 1993