ª
Oncogene (2002) 21, 5716 – 5724
2002 Nature Publishing Group All rights reserved 0950 – 9232/02 $25.00
www.nature.com/onc
Killing of leukemic cells with a BCR/ABL fusion gene by RNA interference
(RNAi)
Monika Wilda1,2, Uta Fuchs1,2, Wilhelm Wössmann1,2 and Arndt Borkhardt*,1
1
Department of Pediatric Hematology and Oncology, University of Giessen, 35392 Giessen, Germany
Short 21-mer double-stranded RNA (dsRNA) molecules
have recently been employed for the sequence-specific
silencing of endogenous human genes. This mechanism,
called RNA interference (RNAi), is extremely potent and
requires only a few dsRNA molecules per cell to silence
homologous gene mRNA expression. We used dsRNA
targeting the M-BCR/ABL fusion site to kill leukemic
cells with such a rearrangement. Transfection of dsRNA
specific for the M-BCR/ABL fusion mRNA into K562
cells depleted the corresponding mRNA and the M-BCR/
ABL oncoprotein. This was demonstrated by real-time
quantitative PCR and Western blots. The BCR/ABL
knockdown was accompanied by strong induction of
apoptotic cell death. Leukemic cells without BCR/ABL
rearrangement were not killed by M-BCR/ABL-dsRNA.
In addition, to corroborate the extraordinary sequence
specificity of RNAi, we designed another RNA oligo
matching the M-BCR/ABL fusion site but having two
point mutations within its central region. We show that
these two point mutations abolished both p210 reduction
and induction of apoptosis in K562 cells. Finally, we
compared leukemic cell killing by RNAi to that caused by
the ABL kinase tyrosine inhibitor, STI 571, Imatinib. For
full induction of apoptosis, dsRNA targeting M-BCR/
ABL required 24 h more than Imatinib. This may be
caused by the relatively long half-life of the BCR/ABL
oncoprotein, which is not targeted by the RNAi
mechanism, but is affected by STI 571. When we applied
ds M-BCR/ABL RNA and STI 571 in combination, we
did not observe a further increase in the induction of
apoptosis. Nevertheless, these data may open a field for
further studies towards gene-therapeutic approaches using
RNA interference to kill tumor cells with specific genetic
abnormalities.
Oncogene (2002) 21, 5716 – 5724. doi:10.1038/sj.onc.
1205653
Keywords: RNA interference; BCR/ABL rearrangement; apoptosis; STI 571; gene therapy
*Correspondence: A Borkhardt, Pediatric Hematology and Oncology, Feulgenstr.12, 35392 Giessen, Germany;
E-mail: Arndt.Borkhardt@paediat.med.uni-giessen.de
2
These authors contributed equally to this work
Received 9 March 2002; revised 9 May 2002; accepted 10 May
2002
Introduction
Chromosomal translocations are the hallmark of
many leukemias, lymphomas, and to a somewhat
lesser extent, solid tumors. They frequently lead to the
generation of chimeric fusion oncoproteins that trigger
malignant transformation. Gene targeting of the
chimeric fusion is an ideal way to kill the tumor cells
specifically, while leaving the normal cells unaffected.
In this light, antisense approaches using singlestranded molecules of either RNA or DNA are
relatively straightforward techniques for these
purposes. However, these attempts have consistently
suffered from some profound problems, e.g. questionable specificity or rather low efficacy (Stein and
Narayanan, 1994; Skorski et al., 1994; Smetsers et
al., 1994). It has recently been discovered that short
dsRNA molecules (siRNAs) trigger a complex
mechanism, called RNA interference, that results in
specific gene silencing even in human cells (Elbashir et
al., 2001a). siRNAs are approximately 21 nucleotides
in length and have a base-paired structure with 2nucleotide
3’-overhang.
Chemically
synthesized
siRNAs may become powerful reagents for genomewide analysis of mammalian gene function in cultured
somatic cells. Beyond their value for target validation,
siRNAs also hold great potential as gene-specific
therapeutic agents. The remarkable potency of the
RNAi reaction enables a complete ‘knock-down’ of a
specific protein and may therefore overcome some of
the initial problems with conventional antisense
strategies. Herein, we explored the feasibility of the
RNAi reaction against an oncogenic fusion gene by
targeting the M-BCR/ABL mRNA in K562 cells
(Figure 1). The BCR/ABL rearrangement is the
molecular equivalent of the Philadelphia chromosome,
which arises from a reciprocal translocation between
chromosomes 9 and 22 (Rowley, 1973). The translocation fuses two unrelated genes, BCR from
chromosome 22 and ABL from chromosome 9, to
form an oncogenic hybrid gene. With regard to the
BCR gene, there are two breakpoint sites, M-BCR
versus m-BCR, and hence two slightly different BCR/
ABL oncoproteins of 210 kD or 190 kD, respectively,
are produced (Chan et al., 1987; Rubin et al., 1988;
Hooberman et al., 1989). The resulting BCR/ABL
protein has, as compared to the normal ABL protein,
an increased kinase activity leading to pathological
Killing of leukemic cells by RNA
M Wilda et al
5717
Figure 1 (a) Schematic representations of the BCR/ABL hybrid gene. The gray colored zones within ABL, or BCR/ABL represent
the area in which the taqman primers/probe are located. The region of the BCR/ABL fusion that was targeted by the dsRNA is
indicated (upper case letters, BCR part; lower case letters ABL part). (b) Schematic structure of the dsRNA that cleaves the target
mRNA. Sense and antisense sequence of both RNA strands that were annealed to dsRNA are shown. To control the specificity of
the RNAi reaction, we additionally designed a siRNA oligo in which two base-pairs were mutated (lower case letters in italics, underlined). These two point mutations do not match within the sequence around the M-BCR/ABL fusion site. (c) Control of transfection efficiency by a FITC-labeled dsRNA. The siRNA was 5’-labeled with FITC and used for transfection. FITC-positive cells were
visualized by fluorescence microscopy and calculated after counter-staining with DAPI
phosphorylation of several downstream targets (Lugo
et al., 1990; Goldman and Druker, 2001a). This
results in oncogenic growth and inhibition of
apoptosis (Druker et al., 1996). The expression of
the BCR/ABL oncoprotein induces a disease resembling CML in mice (Daley et al., 1990). Cytogenetic
and molecular studies of clinical samples revealed that
the rearrangement is found in almost all patients with
chronic myeloid leukemia (CML) and in approximately 30% of adults with acute lymphoblastic
leukemia (ALL) (Dobrovic et al., 1991; Maurer et
al., 1991; Westbrook et al., 1992). The latter subgroup
usually has a poor response to conventional
chemotherapy protocols and thus carries a dismal
prognosis (Lestingi and Hooberman, 1993). However,
we decided to target this particular rearrangement not
only because of its high frequency and paramount
prognostic importance. The small-molecule drug STI
571, now also known as Imatinib, inhibits the
deregulated protein kinase ABL in Ph+ patients. It
has dramatically improved the therapy for BCR/ABLpositive leukemias. Administered orally once daily,
Imatinib had significant anti-leukemic effects even in
patients in whom conventional treatment had failed.
This was accompanied by readily tolerable side effects
(Druker et al., 2001b,c). Experimentally, STI 571
suppressed proliferation of BCR/ABL-expressing cells
and triggered their apoptotic death by various
mechanisms (Druker et al., 1996; Carroll et al.,
1997). In this study, we compared the efficiency of
cell killing by STI 571 to that of BCR/ABL dsRNA in
cells with M-BCR/ABL rearrangement.
Results
Reduction of BCR/ABL mRNA expression by dsRNA
molecules
A prerequisite for the therapeutic application of
siRNAs is that the targeted cells or tissue contain a
functional RNAi mechanism to bind to siRNAs and
mediate mRNA degradation. Original reports about
the successful induction of RNAi in cells of human
origin primarily dealt with HeLa cells (Elbashir et al.,
2001a). In order to test the activity of RNAi in K562
cells, we used a reporter gene assay. Plasmids coding
for firefly and sea-pansy luciferase are co-transfected
together with targeting and control siRNAs, and the
relative luminescence of target and control luciferases is
measured. When thus tested for their ability to
specifically silence luciferase reporters K562 cells,
encouragingly, were responsive to siRNAs (Figure
2a). The same protocol was used to test various
commercially available liposomal transfection reagents
(TransMessenger and Superfect from Quiagen, Hilden,
Germany; Lipofectamine 2000, DMRIE-C, and Oligofectamine from Invitrogen, Paisley, UK) for their
specific ability to deliver plasmid and dsRNA into
Oncogene
Killing of leukemic cells by RNA
M Wilda et al
5718
Figure 2 (a) K562 cells were co-transfected with plasmids and dsRNA as indicated at the bottom of each bar. Cells were subjected
to dual luciferase assay 48 h post-transfection. The luciferase reporter gene regions from plasmids GL3 (firefly) and pRT-TK (renilla) are used according to Elbashir et al., 2001a. The dsRNA Luc targets the firefly luciferase only, while leaving the renilla luciferase unaffected. The ratios of firefly and renilla luciferase are shown. A control panel was transfected with a dsRNA directed at
the human MYC gene (right bars). The average of four independent experiments is shown, error bars indicate standard deviation,
**statistically significant. (b) Example of a representative Lamin amplification plot of the taqman PCR. The Y-axis shows the
threshold above baseline whereas the X-axis shows the number of PCR cycles. Pink curve (A): untreated K562 cells, Yellow (B):
After treatment of K562 cells with dsRNA targeting Lamin, a higher number of PCR cycles is required to pass the fixed threshold
(black horizontal line). (C) Treatment with dsRNA targeting BCR/ABL decreases the number of cycles required to pass the threshold, which indicates that slightly more Lamin mRNA is present in the sample. Please note that these values must be corrected according to the expression data of the housekeeping gene ABL. (c) Absolute copy number of Lamin mRNA per 10 000 copies of ABL
in cells treated with dsRNA targeting BCR/ABL, Lamin or cells without dsRNA treatment. After 48 h, the amount of Lamin
mRNA clearly decreases in the cells transfected with dsRNA against Lamin. (d) Western blot analysis of the Lamin A/C. Note
the slight effect of as RNA. The blot was stripped and re-probed to check for equal loading of total protein
K562 cells. On the basis of our Luciferase assays, we
decided to choose Oligofectamine-based transfections
for the subsequent experiments. Next, we checked
whether an endogenous protein can be downregulated
in K562 cells. We targeted an abundant protein,
Lamin, for which previous studies also have convincingly demonstrated that it can be silenced by dsRNA
in HeLa cells. Using a dsRNA oligo whose sequence
has been published (Elbashir et al., 2001a), we
evaluated whether the knock-down of Lamin can be
reproduced in K562 cells. In addition, we wanted to
calculate the achieved reduction of Lamin mRNA by
quantitative real-time RT – PCR. In order to normalize
for different qualities of input RNA, we used the
expression of the housekeeper ABL gene. The Lamin
mRNA was reduced to 22% of the corresponding
value for the non-transfected controls (Figure 2b,c).
Oncogene
The cleavage of Lamin mRNA was accompanied by
significant reduction of Lamin A/C protein as shown
by Western blotting 48 h after transfection (Figure 2d).
Thus, in K562 cells gene silencing is possible for both
exogenously introduced and endogenously expressed
transcripts.
We next targeted the M-BCR/ABL mRNA and
assayed its expression by taqman PCR. The BCR/ABL
mRNA molecules were quantified and normalized to
10 000 molecules of housekeeper mRNA. To ensure
that our quantitative PCR assay is not prone to
artifacts, we used the expression levels of various
housekeeper genes, the PBGD, the HPRT, the GUS
and the TBP gene (Figure 3a). In contrast to the
quantification procedure for Lamin mRNA, the
expression of normal ABL mRNA was unsuitable for
a housekeeper mRNA. The reason is that the taqman
Killing of leukemic cells by RNA
M Wilda et al
5719
Figure 3 (a) Copies of BCR/ABL mRNA per 10 000 copies of housekeeper mRNA (TBP, PBGD, GUS and HPRT) 48 h posttransfection. The BCR/ABL mRNA is reduced regardless of which housekeeper was used for normalization of RNA input. (b) Western blot analysis, cells were transfected with RNA as indicated at the bottom of each lane. In cells transfected with M-BCR/ABL
dsRNA, the p210 was barely visible but normal p145 ABL is not affected
ABL primer/probes are located in the region that is
present in, and will equally amplify from, the BCR/
ABL fusion gene. Thus, siRNAs transfected would
affect M-BCR/ABL mRNA as well as the normal ABL
control RNA, making comparisons very difficult. As
was found for Lamin mRNA, transfection of siRNA
targeting the M-BCR/ABL fusion site downregulates
the BCR/ABL expression, but to a somewhat lesser
Oncogene
Killing of leukemic cells by RNA
M Wilda et al
5720
extent. The exact copy numbers of M-BCR/ABL in the
various experimental conditions are given in Figure 3a.
In general, the reduction of BCR/ABL mRNA was
observed regardless of whether PBGD, HPRT, GUS or
TBP was used for normalization of input RNA.
Knockdown of M-BCR/ABL protein by dsRNA
Next, we examined whether the p210 BCR/ABL
protein is silenced in K562 cells, which would
correspond to the significant reduction of the MBCR/ABL mRNA after transfection of a 21-mer
dsRNA. As expected, p210 was reduced to an almost
undetectable level in Western blots, whereas neither the
wild-type ABL protein nor the Vimentin was influenced by the dsRNA M-BCR/ABL. To ensure
sequence specificity of the M-BCR/ABL dsRNA we
designed a dsRNA oligo, the sequence of which did not
perfectly match the M-BCR/ABL fusion site. Specifically, we mutated the stretch of four adenosines in the
targeted region from AAAA to ggAA (see Figure 1).
This mutated siRNA was unable to reduce the p210
level, indicating the extraordinary sequence specificity
of RNAi (Figure 3b). We finally wanted to test RNAiapproach for the m-BCR/ABL fusion as well.
Unfortunately, we failed with various attempts to
efficiently transfect SD-1 cells that display this minor
fusion transcript (data not shown).
Induction of apoptosis by ds M-BCR/ABL RNA and
STI 571
Previous studies revealed that downregulation of BCR/
ABL renders K562 cells susceptible to induction of
apoptosis by chemotherapeutic agents (McGahon et
al., 1994). We looked for the induction of apoptosis
48 h and 72 h after transfection. As summarized in
Figure 4a, 48 h after transfection with ds M-BCR/ABL
the rate of apoptosis in K562 cells was above that in
the controls but did not reach the same level as in the
STI 571-treated cells. Twenty-four hours later,
however, the number of Histone-associated DNA
fragments had become the same in K562 cells treated
with 1 mM STI 571 as in cells transfected with ds MBCR/ABL. In contrast, single-stranded antisense MBCR/ABL did not induce apoptosis above the control
level. Perhaps not surprisingly, in our assay we did not
see an additive effect when STI 571 and dsRNA were
combined. To provide further evidence that neither the
dsRNA itself nor the transfection reagent induces
apoptosis, we transfected a series of cell lines (HeLa,
293, and Su-DHL) that do not contain a M-BCR/ABL
rearrangement. In none of these cells was apoptosis
induced (data not shown). When K562 cells were
treated with a dsRNA that spans the m-BCR/ABL
fusion, we also did not detect apoptotic cell death
(Figure 4a). Finally, K562 cells that were transfected
with the M-BCR/ABL siRNA having the two point
mutations also failed to show apoptosis. These ELISA
data were thus in good accordance with the lack of
p210 reduction. Morphologically, we saw membrane
Oncogene
vacuolization and destruction in 82 or 57% of the
K562 cells that were treated with 1 M STI 571 or
transfected with dsRNA M-BCR/ABL, respectively.
Again, cells transfected with either single-stranded
asRNA against M-BCR/ABL or ds-m-BCR/ABL
showed no signs of apoptosis above the control level
(Figure 4b).
Discussion
In patients with leukemia and translocation t(9;22) or
the BCR/ABL rearrangement, the molecular-targeted
tumor therapy has dramatically improved through the
development of the small-molecule drug, STI 571
(Goldman and Druker, 2001a; Druker et al., 2001a).
It has considerable advantages over conventional
treatment modalities (interferon alpha), e.g. rapid and
more frequent hematological and cytogenetic responses
combined with fewer side effects. In recent studies,
several authors reported the development of resistance
to STI 571, e.g. by genomic amplification of BCR/
ABL, increased expression of BCR/ABL mRNA or
point mutation in the ABL gene (Mahon et al., 2000;
Gorre et al., 2001; Hochhaus et al., 2001; Barthe et al.,
2001). Thus, it is currently a common belief that STI
571 alone cannot cure CML or Ph+ positive ALL,
and that the development of additional therapeutic
approaches would be of interest for those patients
(Goldman and Melo, 2001b). Towards this end, we
have shown here that the specific silencing of BCR/
ABL mRNA by dsRNA-induced RNA interference is
nearly as effective as STI 571 in tumor-cell killing. This
approach, however, does not affect the oncoprotein
itself but rather its corresponding fusion mRNA. In the
light of the relatively long half-life of the BCR/ABL
protein (Dhut et al., 1990) it seems understandable that
the cells were killed less rapidly than they are by STI
571. The cell killing seen in our experiments can clearly
be attributed to an RNAi effect, since single-stranded
antisense RNA did not affect cell viability (Figure 4).
In the past, other studies used antisense DNA directed
against the BCR/ABL fusion and were also able to
demonstrate an impressive reduction of mRNA but,
unfortunately, not of p210 BCR/ABL. This discrepancy was due to the fact that the antisense effect was
only transient and suppression of BCR/ABL mRNA
was wearing off after 8 h. Thus, BCR/ABL was
considered to be a ‘difficult oncoprotein to target’
(Spiller et al., 1998) by antisense approaches. RNAi is
by far more potent and enables the induction of the
phenotype ‘apoptosis’ and not of a M-BCR/ABL
mRNA or p210 reduction only. Given that the
transfection rates were only 80%, the achieved BCR/
ABL mRNA reduction is quite impressive. A difference
of 3.3 Ct values (threshold cycles) corresponds to
approximately one order of magnitude. Thus, three
PCR cycles later, with only 10% of non-transfected
cells within the dsRNA treated population, the amount
of PCR product will be similar to that in a population
comprising exclusively untreated cells. However, this is
Killing of leukemic cells by RNA
M Wilda et al
5721
Figure 4 (a) Apoptosis in K562 cells treated with ds M-BCR/ABL, ds-M-BCR/ABL mut., antisense or sense BCR/ABL RNA as
well as with STI 571. To further ensure sequence specificity of the RNAi effect, we transfected dsRNA corresponding to the mBCR/ABL fusion that did not induce apoptosis. The targeted region of m-BCR/ABL was 5’-AUGGAGACGCAGAAGCCCTT3. In addition, M-BCR/ABL dsRNA having two point mutations in its central region also lacks the apoptosis-inducing effect.
The average of five independent experiments is shown, error bars indicate standard deviation, statistically significant increase of
apoptosis above the controls (*P50.05), (**P50.01). (b) Morphology of K562 cells 48 h after transfection. Extensive vacuolization
was seen in K562 cells treated with STI 571 and to a somewhat lesser extent in cells transfected with dsRNA targeting M-BCR/ABL
but not in cells treated with either ds m-BCR/ABL or as m-BCR/ABL
valid only if the 90% transfected cells show a complete
BCR/ABL mRNA knockdown and are totally free of
BCR/ABL mRNA.
Another limitation of RNAi-targeting experiments is
the transient nature of RNA transfer and the
requirement for synthesis of RNA oligos before
application (Tuschl, 2002). The intracellular expression
of double-stranded RNA molecules from plasmid
DNA is an attempt to suppress this limitation.
Brummelkamp et al. (2002) used this approach to
produce cells that stably suppress p53 over a period of
2 months. Incorporation of dsRNA expression
Oncogene
Killing of leukemic cells by RNA
M Wilda et al
5722
cassettes into alternative vector systems, e.g. retroviral
vectors, may also pave the way for targeting primary
cells previously refractory to dsRNA treatment by
liposomal transfection methods.
However, one should keep in mind that the sequencespecific mRNA degradation by the RNAi is an active
process that requires the proper function of a complex
network of endogenous proteins. Whether the intrinsic
ability to use the RNAi machinery is preserved in all
cancer cells or whether cancer cells may rapidly develop
a resistance to this therapeutic mRNA degradation
should be addressed by future studies. One biochemical
antagonistic effect to RNAi has recently been found
(Scadden and Smith, 2001). The authors analysed the
human adenosine deaminase that acts on dsRNA,
ADAR2. In their studies, RNAi was inhibited when
the dsRNA molecule was first deaminated by ADAR2.
It is tempting to speculate that tumor cells may use this
defense mechanism and upregulate such enzymes in
response to therapeutic interventions by dsRNA. Our
study further demonstrates that siRNAs are highly
sequence-specific reagents and discriminate between
mismatched target RNA sequences (Elbashir et al.,
2001b). This predicts an additional means by which
tumor cells may escape a therapeutic RNAi intervention:
simply by mutation of the target fusion site. Nevertheless, during the last 2 decades the combined effort of
many laboratories worldwide has led to the molecular
clarification of numerous chromosomal translocations
by cloning the genes involved (Rabbitts, 1994, 1998;
Rowley, 1999). Silencing of these tumor-specific
chimeric mRNAs by RNAi, as exemplified here using
the M-BCR/ABL rearrangement, may become a promising new approach towards a molecularly targeted tumor
therapy. In the near future, exploration of the RNAimediated gene therapy in mouse models (Corral et al.,
1996) will provide helpful information as to whether
RNAi-mediated gene therapy can really become translated into patient therapy.
according to the manufacturer’s instructions. To avoid
contamination with E. coli RNA, we digested the resulting
plasmid DNA with DNAse-free RNAse for 1 h at 378C
(Roche Diagnostics, Mannheim, Germany). The concentration of plasmid DNA was measured spectrophotometrically
and copy numbers were determined according to the
molecular weight of the respective inserts (for insert size see
Table 1). One representative of each standard plasmid was
sequenced to exclude misincorporation of single nucleotides
by taq-polymerase. The plasmid standards can be obtained
upon request.
Quantitative PCR
In the taqman PCR (taqman 7700, Perkin Elmer, Foster
City, CA, USA) reactions are characterized by the point
during cycling when the PCR product is first detected (the
threshold cycle, Ct) rather than the amount of PCR product
accumulated after a fixed number of cycles. The amounts of
the various target messages, e.g. BCR/ABL, ABL, PBGD,
HPRT, Lamin, and TBP were quantified by measuring Ct
and by using a standard curve to determine the starting
target message quantity. One principal problem for quantification of mRNA by plasmid standards is that there may be
variance within the reverse transcription (RT) reaction that is
not monitored during the procedure. Thus, we carefully
assessed the efficiency of the cDNA synthesis by calculating
the amount of cDNA after its synthesis in a set of separate
experiments. Quantification of cDNA was done by Oligreen
(Molecular Probes, Leiden, The Netherlands), which binds to
single-stranded DNA only. When our protocol for cDNA
synthesis (see below) is used, 90 – 95% of all input RNA
molecules are converted into cDNA after 1 h (D Rawer,
personal communication). This value was very stable and did
not vary between the different target genes.
For the generation of the external standard curve, we
diluted the plasmid DNA in 10-fold steps, giving a range of
10 – 106 molecules. The correlation coefficients between the
threshold cycle and the starting quantity of the various
standard DNAs were around 0.99. Furthermore, the slope of
the standard curves nearly matched the theoretical value of
73.33, (data not shown). Quantification was performed in
duplicate and we observed a minimal intra-assay variation for
each sample, corresponding to per cent variance of copy
numbers between 3.8 and 7.9%.
Materials and methods
RNA isolation and cDNA synthesis
Cell culture and treatment with STI 571
K562 cells were obtained from the German Collection of
Microorganisms and Cell Cultures (DMSZ, Braunschweig,
Germany, http://www.dsmz.de). Cells were routinely maintained in RPMI1640 medium supplemented with 10% fetal
calf serum (FCS) without antibiotics in a humidified atmosphere of 5% CO2 at 378C. STI 571, Imatinib, was kindly
provided by Novartis (Novartis, Switzerland). It was added
at a concentration of 1 mM to exponentially growing cells.
Generation of PCR standards
For absolute quantification of template copy number we first
cloned cDNA fragments of Lamin, M-BCR/ABL, PBGD,
HPRT and the TATA-box binding protein (TBP) into the
pCR II TOPO plasmid (Invitrogen, Groningen, The Netherlands). Primers used for generation of standards are shown in
Table 1. Plasmids from single colonies were prepared with
ion chromatography columns (PeqLab, Erlangen, Germany)
Oncogene
The RNA was isolated by means of a standard protocol with
guanidium thiocyanate phenol-chloroform. For cDNA synthesis, we used a modified protocol which ensures that almost
all RNA is converted into cDNA. Five hundred ng of total
RNA was incubated with 100 ng oligo dT primers (Roche,
Diagnostics), 1000 U Superscript II (Invitrogen), and 5 ml
dNTP (10 nM), for 10 min at 258C followed by 50 min at
458C and 15 min at 708C. Reactions were carried out in a
final volume of 100 ml containing the buffer supplied by the
manufacturer (Invitrogen). As stated above, quantification of
cDNA after the reverse transcription step was performed
with Oligreen (Molecular Probes, Leiden, Germany) and the
result was compared with the input RNA measured spectrophotometrically.
Housekeeper genes for quantitative PCR
We selected five housekeeper genes as endogenous RNA
control and the samples were normalized on the basis of their
Killing of leukemic cells by RNA
M Wilda et al
5723
Table 1 Forward (FP), reverse (RP) primers and probes used for taqman PCR
Target
gene
PCR
product bp
Accession
Real-Time quantitative taqman PCR
M
BCR/ABL
AJ 131466
ABL
AJ 131466
TBP
NM 003194
Lamin
XM 002071
HPRT
NM 000194
PBGD
NM 000190
Probe 5’-3’: agcccttcagcggccagtagcatc
FP 5’-3’: cgtccactcagccactggat
RP 5’-3’: agttccaacgagcggcttc
Probe 5’-3’: caacaccctggccgagttggttcat
FP 5’-3’: caacactgcttctgatggcaa
RP 5’-3’: cggccaccgttgaatgat
Probe 5’-3’: actgttcttcactctcttggctcctgtgca
FP 5’-3’: gcatattttcttgctgccagtct
RP 5’-3’: accacggcactgattttcagtt
Probe 5’-3’: gcttggtctcacgcagctcctcactgta
FP 5’-3’: aatgatcgcttggcggtcta
RP 5’-3’: aggttgctgttcctctcagcag
Probe 5’-3’: ccatgttcaattatatcttccacaatcaagac
FP 5’-3’: aggaaagcaaagtctgcattgtt
RP 5’-3’: ggtggagatgatctctcaactttaa
Probe 5’-3’: ctgttttcttccgccgttgcagc
FP 5’-3’: cccacgcgaatcactctcat
RP 5’-3’: tgtctggtaacggcaatgcg
104 bp
Standard
Calibrator
clone, bp
insert
FP 5’-3’: tcacggatctcagcttccagatgg
RP 5’-3’: ttgtgcttcatggtgatgtccgtg
1861 bp
92 bp
See M-BCR-ABL
1861 bp
90 bp
FP 5’-3’: cactgtttcttggcgtgtgaa
RP 5’-3’: aaccaggaaataactctggctcata
1016 bp
110 bp
FP 5’-3’: gcatcaccgagtctgaagaggt
RP 5’-3’: tcccattgtcaatctccaccag
94 bp
FP 5’-3’: aggaaagcaaagtctgcattgtt
RP 5’-3’: ggtggagatgatctctcaactttaa
71 bp
FP 5’-3’: aacggtggtgtgacaggcag
RP 5’-3’: tgtctggtaacggcaatgcg
512 bp
94 bp
120 bp
For all probes TAMRA and FAM fluorescent dyes were used as quencher or reporter, respectively. The standard plasmids were generated with
the primers shown on the right side. The b-Glucoronidase gene (GUS, Accession number NM_000181) used for normalization of M-BCR/ABL
copy number was amplified with the ‘ready to use’ pre-developed assay from Applied Biosystems (ABI). ABI does not provide its customers
with the sequences of either primers or probe
housekeeper content. The housekeeper RNA was also
quantified by a standard plasmid curve. We rejected several
commonly used housekeeper genes, such as b-actin, b-2
micoglobulin, and 18 S RNA, for several reasons, e.g. the
existence of pseudogenes, the lack of introns or very high
abundance of transcripts. Instead, we used ABL, HPRT,
GUS, PDBP, and TBP, a component of the DNA-binding
protein complex TFIID. To ensure RNA specificity of the
taqman PCR, all primer/probe combinations were positioned
over exon/intron boundaries. We then tested all primer/probe
combinations using genomic DNA as template and did not
observe an amplification product after 40 cycles of PCR. The
taqman PCR was performed according to published protocols of the manufacturer (see http://docs.appliedbiosystems.
com/).
in hypotonic cell lysis buffer +350 mM NaCl, lysed by
pipetting and incubated for 10 min on ice. The samples were
diluted 1 : 2 with non-reducing sample buffer (1 ml: 60 ml 1 M
Tris-HCl pH 6.8; 312 ml 80% Glycerol; 200 ml 10% SDS;
428 ml H2O; grains of Bromphenol blue) and electrophoresed
on an 8% SDS-polyacrylamide gel.
The antibodies were commercially obtained from Santa
Cruz Biotechnology Inc. (Santa Cruz). In a standard Western
blot protocol, for detection of the BCR/ABL fusion proteins
and the ABL wild-type protein we used a rabbit polyclonal
antibody against the C-terminus of c-ABL (C-19). Lamin A/
C and Vimentin antibodies were used as described previously
(Elbashir et al., 2001a). The protein was detected by
chemoluminescence, by means of the ECL system (Amersham, Uppsala, Sweden).
Source of dsRNA molecules, transfection and luciferase assay
Detection of apoptosis
The dsRNA’s were commercially obtained from dharmacon
(Lafayette, Co. USA). For transfection, either dsRNA or
ssRNA molecules were handled exactly according to the
procedure used by the Tuschl laboratory (http://www.
mpibpc.gwdg.de /abteilungen / 100 / 105/ siRNAuserguide.pdf),
which has also been distributed by dharmacon and summarized
in their user manual (http://www.dharmacon.com/sirna.html).
The dsRNA sequence for silencing the Lamin A/C mRNA and
the firefly luciferase reporter gene region were used according to
the work published by the Tuschl group (Elbashir et al., 2001a).
The sequence of the M-BCR/ABL dsRNA as well as the oligo
with two point mutations are shown in Figure 1c. Expression of
firefly and sea-pansy luciferase was monitored with the Dual
luciferase kit according to the manufacturer (Promega,
Madison, USA) in a Berthold Luminometer (LB953, Bad
Wildbad, Germany).
Cells were washed with PBS (pH 7.3), resuspended and
examined as cytospin preparations. We evaluated the cells
morphologically in the light microscope after Wright staining.
In all, five different fields were randomly selected for counting
200 cells. The percentage of apoptotic cells was calculated
(Ray et al., 1994). Histone-associated DNA-fragments in the
cytoplasmic fraction of cell lysates were detected by means of
a sandwich-ELISA purchased from Roche (Roche Diagnostics, Mannheim, Germany). The cytoplasmic fractions of cell
lysates from 56103 cells were incubated for 2 h with
biotinylated antibodies directed against histones and peroxidase-coupled anti-DNA-antibodies. After removal of
unbound antibodies, ABTS was added as a peroxidasesubstrate. Absorption was measured at 405 nm.
Cell extraction and Western blotting
Cell samples were centrifuged at 48C in a microfuge to pellet
the nuclei. For nuclear extracts, the nuclei were resuspended
Statistical analysis
Both, values of Luciferase expression in variously transfected
K562 cells and the rate of apoptosis therein were analysed by
U-test according to Mann and Whitney. P values 50.05 were
considered significant.
Oncogene
Killing of leukemic cells by RNA
M Wilda et al
5724
Acknowledgments
The authors wish to thank Dr T Tuschl, Dr S Viehmann
and D Rawer for their help with the design of the dsRNA,
quantification of BCR/ABL mRNA by taqman PCR, or
generation of plasmid standards, respectively. Expert
technical assistance by Stefanie Garkisch is gratefully
acknowledged. A part of these studies was supported by
a grant from the Deutsche Forschungsgemeinschaft to A
Borkhardt.
References
Barthe C, Cony-Makhoul P, Melo JV and Mahon JR. (2001).
Science, 293, 2163.
Brummelkamp TR, Bernards R and Agami R. (2002).
Science, 296, 550 – 553.
Carroll M, Ohno-Jones S, Tamura S, Buchdunger E,
Zimmermann J, Lydon NB, Gilliland DG and Druker
BJ. (1997). Blood, 90, 4947 – 4952.
Chan LC, Karhi KK, Rayter SJ, Heisterkamp N, Eridani S,
Powles R, Lawler SD, Groffen J, Foulkes JG, Greaves MF
and Wiedemann LM. (1987). Nature, 325, 635.
Corral J, Lavenir I, Impey H, Warren AJ, Forster A, Larson
TA, Bell S, McKenzie AN, King G and Rabbitts TH.
(1996). Cell, 85, 853 – 861.
Daley GQ, van Etten RA and Baltimore D. (1990). Science,
247, 824 – 830.
Dhut S, Chaplin T and Young BD. (1990). Leukemia, 4,
745 – 750.
Dobrovic A, Morley AA, Seshadri R and Januszewicz EH.
(1991). Leukemia, 5, 187 – 190.
Druker BJ, Sawyers CL, Capdeville R, Ford JM, Baccarani
M and Goldman JM. (2001a). Hematology (Am. Soc.
Hematol. Educ. Program.) 87 – 112.
Druker BJ, Sawyers CL, Kantarjian H, Resta DJ, Reese SF,
Ford JM, Capdeville R and Talpaz M. (2001b). N. Engl. J.
Med., 344, 1038 – 1042.
Druker BJ, Talpaz M, Resta DJ, Peng B, Buchdunger E,
Ford JM, Lydon NB, Kantarjian H, Capdeville R, OhnoJones S and Sawyers CL. (2001c). N. Engl. J. Med., 344,
1031 – 1037.
Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM,
Fanning S, Zimmermann J and Lydon NB. (1996). Nat.
Med., 2, 561 – 566.
Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K
and Tuschl T. (2001a). Nature, 411, 494 – 498.
Elbashir SM, Martinez J, Patkaniowska A, Lendeckel W and
Tuschl T. (2001b). EMBO J., 20, 6877 – 6888.
Goldman JM and Druker BJ. (2001a). Blood, 98, 2039 –
2042.
Goldman JM and Melo JV. (2001b). N. Engl. J. Med., 344,
1084 – 1086.
Gorre ME, Mohammed M, Ellwood K, Hsu N, Paquette R,
Rao PN and Sawyers CL. (2001). Science, 293, 876 – 880.
Hochhaus A, Kreil S, Corbin A, La Rosee P, Lahaye T,
Berger U, Cross NC, Linkesch W, Druker BJ, Hehlmann
R, Gambacorti-Passerini C, Corneo G and D’Incalci M.
(2001). Science, 293, 2163.
Oncogene
Hooberman AL, Carrino JC, Leibowitz D, Rowley JD, Le
Beau MM, Arlin Z and Westbrook CA. (1989). Proc. Natl.
Acad. Sci. USA, 86, 4259.
Lestingi TM and Hooberman AL. (1993). Hematol. Oncol.
Clin. North Am., 7, 161 – 175.
Lugo TG, Pendergast AM, Müller AJ and Witte ON. (1990).
Science, 247, 1079 – 1082.
Mahon FX, Deininger MW, Schultheis B, Chabrol J,
Reiffers J, Goldman JM and Melo JV. (2000). Blood, 96,
1070 – 1079.
Maurer J, Janssen JWG, Thiel E, Van Denderen J, Ludwig
WD, Aydemir Ü, Heinze B, Fonatsch C, Harbott J, Reiter
A, Riehm H, Hoelzer D and Bartram CR. (1991). Lancet,
337, 1055 – 1058.
McGahon A, Bissonnette R, Schmitt M, Cotter KM, Green
DR and Cotter TG. (1994). Blood, 83, 1179 – 1187.
Rabbitts TH. (1994). Nature, 372, 143 – 149.
Rabbitts TH. (1998). N. Engl. J. Med., 338, 192 – 194.
Ray S, Ponnathpur V, Huang Y, Tang C, Mahoney ME,
Ibrado AM, Bullock G and Bhalla K. (1994). Cancer
Chemother. Pharmacol., 34, 365 – 371.
Rowley JD. (1973). Nature, 243, 290 – 293.
Rowley JD. (1999). Semin. Hematol., 36, 59 – 72.
Rubin CM, Carrino JJ, Dickler MN, Leibowitz D, Smith SD
and Westbrook CA. (1988). Proc. Natl. Acad. Sci. USA,
85, 2795 – 2799.
Scadden AD and Smith CW. (2001). EMBO Rep., 2, 1107 –
1111.
Skorski T, Nieborowska-Skorska M, Nicolaides NC,
Szczylik C, Iversen P, Iozzo RV, Zon G and Calabretta
B. (1994). Proc. Natl. Acad. Sci. USA, 91, 4504 – 4508.
Smetsers TF, Skorski T, van de Locht LT, Wessels HM,
Pennings AH, de Witte T, Calabretta B and Mensink EJ.
(1994). Leukemia, 8, 129 – 140.
Spiller DG, Giles RV, Grzybowski J, Tidd DM and Clark
RE. (1998). Blood, 91, 4738 – 4746.
Stein CA and Narayanan R. (1994). Curr. Opin. Oncol., 6,
587 – 594.
Tuschl T. (2002). Nat. Biotechnol., 20, 446 – 448.
Westbrook CA, Hooberman AL, Spino C, Dodge RK,
Larson RA, Davey F, Wurster-Hill DH, Sobol RE,
Schiffer C and Bloomfield CD. (1992). Blood, 80, 2983 –
2990.