FEBS 28595 FEBS Letters 570 (2004) 195–204

Suppression of bcr-abl synthesis by siRNAs or tyrosine kinase activity

by Glivec alters different oncogenes, apoptotic/antiapoptotic genes
and cell proliferation factors (microarray study)
Zhivko Zheleva, Rumiana Bakalovaa,*, Hideki Ohbaa, Ashraf Ewisa, Mitsuru Ishikawaa,
Yasuo Shinoharaa,b, Yoshinobu Babaa,b
a
Single-Molecule Bioanalysis Laboratory, National Institute for Advanced Industrial Science and Technology, AIST-Shikoku,
2217-14 Hayashi-cho, Takamatsu, Kagawa 761-0395, Japan
b
Faculty of Pharmaceutical Sciences, Tokushima University, Tokushima, Japan

Received 15 March 2004; revised 16 June 2004; accepted 17 June 2004

Available online 1 July 2004

Edited by Varda Rotter

activity that inhibits apoptosis and triggers malignant trans-

Abstract Short 21-mer double-stranded/small-interfering
RNAs (ds/siRNAs) were designed to target bcr-abl mRNA in formation. Bcr-abl fusion is a single causative abnormality in
chronic myelogenous leukemia. The ds/siRNAs were transfected chronic myelogenous leukemia (CML), at least in chronic
into bcr-abl-positive K-562 (derived from blast crisis chronic phase, making it a unique model for development of molec-
myelogenous leukemia), using lipofectamine. Penetrating of ular targets for disease control [1–3]. Gene or protein tar-
ds/siRNAs into the cells was detected by fluorescent confocal geting of the chimeric fusion is a promising way to eliminate
microscopy, using fluorescein-labeled ds/siRNAs. The cells were the bcr-abl-positive cells specifically, without any influence on
treated with mix of three siRNA sequences (3
60 nM) during 6 the viability of normal cells. Increasing knowledge on bcr-abl
days with three repetitive transfections. The siRNA-treatment resulted in a design of several novel therapeutic approaches,
was accompanied with significant reduction of bcr-abl mRNA, including highly specific tyrosine kinase inhibitors (as Glivec
p210, protein tyrosine kinase activity and cell proliferation
and pyrido[2,3-d]pyrimidine derivative PD180970), single-
index. Treatment of cells with Glivec (during 8 days with four
repetitive doses, 180 nM single dose) resulted in analogous stranded antisense oligo-RNA and oligo-DNA substances,
reduction of cell proliferation activity, stronger suppression of and immunomodulation [1–4]. Recently, it has been reported
protein tyrosine kinase activity, and very low reduction of p210. that short double-stranded/small-interfering RNA molecules
siRNA-mix and Glivec did not affect significantly the viability of (ds/siRNA), targeting to bcr-abl abnormality, are also
normal lymphocytes. Microarray analysis of siRNA- and Glivec- promising tool for control of the chimeric fusion protein ex-
treated K-562 cells demonstrated that both pathways of bcr-abl pression [5–7].
suppression were accompanied with overexpression and suppres- It has been found that the inhibition of bcr-abl synthesis by
sion of many different oncogenes, apoptotic/antiapoptotic and antisense oligo-DNAs or conventional drugs is accompanied
cell proliferation factors. The following genes of interest were with cross-talk with other oncogenes (e.g., telomerase) and
found to decrease in relatively equal degree in both siRNA- and
telomere-associated proteins (e.g., tankyrase, TRF1 and Tin2)
Glivec-treated cells: Bcd orf1 and orf2 proto-oncogene, chroma-
tin-specific transcription elongation factor FACT 140-kDa [8,9]. The overexpression of antiapoptotic and other cell pro-
subunit mRNA, gene encoding splicing factor SF1, and mRNA liferation factors, together with point-mutations in abl-domain
for Tec protein tyrosine kinase. siRNA-mix and Glivec provoked [10,11], can be responsible for development of resistance to
overexpression of the following common genes: c-jun proto- the conventional anti-bcr-abl drugs after their long-term
oncogene, protein kinase C-a, pvt-1 oncogene homologue (myc application.
activator), interleukin-6, 1-8D gene from interferon-inducible The discovery of RNA interference (RNAi) mechanism in
gene family, tumor necrosis factor receptor superfamily (10b), mammalian cells in 2001 [12,13] enhanced the expectations to
and STAT-induced STAT inhibitor. decide the problems of gene therapy of bcr-abl abnormality.
Ó 2004 Published by Elsevier B.V. on behalf of the Federation of The efforts were directed to the application of this natural
European Biochemical Societies.
phenomenon for a specific influence of oncogene and onco-
protein expression.
RNAi is conserved in a diverse variety of simpler organisms
(such as nematodes, plants and Drosophila) [14–16]. RNAi is
1. Introduction mediated by siRNAs that are produced from long dsRNAs of
exogenous or endogenous origin by an endonuclease of the
The bcr-abl oncogene is a result of a reciprocal transloca- ribonuclease-III type, called Dicer. The resulting siRNAs are
tion between the long arms of chromosome 9 and 22, en- about 21–23 nucleotides long and these short fragments serve
coding a cytoplasmic fusion oncoprotein with tyrosine kinase as guide sequences to induce target-specific mRNA cleavage by
the RNA-induced silencing complex [17–19]. This mechanism
*
Corresponding author. Fax: +81-87-8694113. is extremely potent and requires only a few dsRNA molecules
E-mail address: [email protected] (R. Bakalova). per cell to silence homologous gene mRNA expression.

0014-5793/$22.00 Ó 2004 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies.
doi:10.1016/j.febslet.2004.06.048
196 Z. Zhelev et al. / FEBS Letters 570 (2004) 195–204

It has been widely accepted that siRNAs are useful tool in Mg2þ free) to eliminate free fluorescent ds/siRNA molecules outside
gene knockdown experiments and ultimately for therapeutic the cells, and the FITC-labeled ds/siRNA, incorporated into the living
purposes because siRNA-mediated transcriptional silencing is cells (green light), were detected by fluorescent confocal microscopy.
The samples were analyzed by Olympus IX70 microscope.
highly gene-specific and without secondary effects. However,
caution is needed because of the findings that some siRNAs 2.4. Treatment with Glivec
induce interferon-stimulated genes in mammalian cells in vitro Glivec was kindly provided by Novartis (Novartis, Switzerland). It
[20]. Moreover, it is too early to say how great is the promise of was applied to the cells in a concentration of 180 nM every two days
(the protocol is similar to that in Fig. 1C) during 8 days to obtain
RNAi because of many queries to the estimation of real
analogous to siRNA-protocol reduction of cell proliferation.
efficacy of siRNAs, as their extremely difficult intracellular
delivery without transfectants, transient nature of RNAi, side- 2.5. mRNA isolation
effects during long-term treatment, etc. mRNA was isolated from ds/siRNA- or Glivec-treated and -un-
The present study was designed to compare the effects of two treated cells, using QuickPrep mRNA Purification kit (Amersham
Pharmacia), as described in the manufacturer’s protocol. Briefly, total
independent pathways for bcr-abl suppression (siRNA-medi- RNA was extracted in a buffered solution containing a high concen-
ated inhibition of bcr-abl protein synthesis and Glivec-medi- tration of GTC and N-lauroyl sarcosine, which ensured the rapid in-
ated inhibition of tyrosine kinase activity of already activation of endogenous RNase and the complete dissociation of
synthesized bcr-abl protein) on the cross-expression of other cellular components from the mRNA. The extracts were then diluted
oncogenes, apoptotic/antiapoptotic genes and cell prolifera- threefold with elution buffer (10 mM Tris–HCl, pH 7.5 and 1 mM
EDTA) to reduce the GTC concentration to a level enough to allow
tion factors that can be responsible for restoration of cell im- efficient hydrogen-bonding between poly(A) tracts on mRNA mole-
mortalization and development of drug resistance. cules and oligo(dT) attached to cellulose, but high enough to maintain
complete inhibition of RNases. Threefold dilution also caused pre-
cipitation of some proteins, which were removed by centrifugation
(12 000
g for 10 min). The supernatant was poured into an oligo(dT)-
2. Materials and methods cellulose spun column and polyadenylated fraction was allowed to
bind over a short period of time with gentle mixing (10 min). The
2.1. siRNAs column was subjected to a low-speed centrifugation (350
g for 2 min)
Twenty one nucleotide ds/siRNAs were chemically synthesized and the liquid, containing the non-bound material, was decanted. The
(Genenet Co. Ltd., Fukuoka, Japan). They were directed to human matrix was batch-washed sequentially with high-salt (10 mM Tris–
bcr-abl fusion peptide (major breakpoint, abl gene). The sense cDNA HCl, pH 7.5, 1 mM EDTA and 0.5 M NaCl) and then low-salt buffer.
sequence was in exon 548  843 (number 3, abl gene) – for siRNA-1 Finally, the sample was eluted from the matrix prewarmed at 65 °C
and siRNA-3; and in exon 374  547 (number 2, abl gene) – for siR- with elution buffer (10 mM Tris–HCl, pH 7.5 and 1 mM EDTA).
NA-2 (GenBank Accession No. AJ131466.1 and X16416.1). The sense Precipitation of mRNA from eluate (0.5 ml) was carried out by
mRNA/cDNA target sequences and antisense siRNA sequences are potassium acetate solution (50 ll–5 M, pH 5.0), glycogen solution
shown in Fig. 1A. The following mismatch ds/siRNA (non-sense) was (10 ll–20 mg/ml in DEPC-treated water) and 95% ethanol (1 ml) for 30
used as a negative control: siRNA-4 – r(GUUUGUCAGAGU min at )20 °C. The precipitated mRNA was resuspended in DEPC-
CGGACAG)d(TT)/r(CUGUCCGACUCUGACAAAC)d(TT). treated water and its concentration was determined spectrophoto-
metrically at 280 nm. The mRNA isolated was essentially free of DNA
2.2. Cells and siRNA transfection protocol and protein contamination. Aliquots of mRNA in equal concentration
Bcr-abl positive K-562 cells (derived from CML patients in blast were applied for reverse transcription-polymerase chain reaction (RT-
crisis) were a generous gift of Dr. J. Minowada (Hayashibara Bio- PCR) analysis.
chemical Laboratories, Inc., Okayama, Japan). The cells were cultured
in RPMI-1640 medium, supplemented with 10% heat-inactivated fetal 2.6. RT-PCR
bovine serum in a humidified atmosphere at 37 °C with 5% CO2 . The RT-PCR was performed by OneStep RT-PCR kit (Qiagene) in ac-
cells used for transfection were about 80% confluenced. The cells used cordance to the manufacturer’s instructions. The following gene-spe-
for assay were in a logarithmic phase. cific primers were used: abl-FP (24-mer), 50 -CTC ATA TCA ACC
Normal lymphocytes were purified from heparinized peripheral CGA GTG TCT CTT-30 ; abl-RP (24-mer), 50 -TGC TAC CTC TGC
blood obtained from normal adults (aged 40–42 years) by Lym- ACT ATG TCC ATG-30 ; bcr-abl fusion-FP (26-mer), 50 -GAA GAA
phosepar I. GTG TTT CAG AAG CTT CTC CC-30 ; bcr-abl fusion-RP (25-mer),
All cells were sedimented by centrifugation (1000 rpm, 10 min) and 50 -GAC CCG GAG CTT TTC ACC TTT AGT T-30 . The reverse-
washed three times by PBS()) before experiments. transcription reaction was carried out at 50 °C for 30 min, followed by
Transfection conditions: collection of cells by centrifugation, re- HotStar Taq DNA polymerase activation by heating at 95 °C for
suspendation of cells in a new medium to 8
105 cells/ml, transfection 15 min. Three-step cycling was performed: denaturation – 1 min at 94
of siRNAs (mix of the three sequences – 3
60 nM) using lipofecta- °C, annealing – 1 min at 50–68 °C, and extension – 1 min at 72 °C. The
mine (6 h incubation in RPMI-1640 medium in the presence of 10 U/ml number of cycles was 40, followed by final extension at 72 °C for
RNase inhibitor (Roche)), replacement of cells in a new medium 10 min. cDNA (RT-PCR product) was dissolved in loading buffer and
(4
105 cells/ml) and cultivation in a humidified atmosphere. The was analyzed by electrophoresis on 1% agarose precast gel with ethi-
conditions for preparation of ds/siRNA–lipofectamine complexes are dium bromide (100 V, 30–40 min, at room temperature, RT  22 °C).
described in Lipofectamine’2000 protocol (Invitrogen). Transfection To ensure that there are no artifacts in our RT-PCR procedure, we
efficacy was analyzed using lamin A/C system (RNAi Started kit, analyzed also the expression levels of housekeeper b-actin mRNA in
Qiagen). Three repetitive transfections were applied as shown in each sample as endogenous reference, using commercially available
Fig. 1C. primers (Qnet).
The half-life of ds/siRNAs in RPMI-1640 medium during transfec-
tion was analyzed by gel electrophoresis (using UV-detection) and
2.7. Western blot analysis of p210 bcr-abl protein
HPLC (using fluorescein-labeled ds/siRNA and fluorimetric detection).
The cells were treated with ds/siRNAs or Glivec as described above.
Aliquots of 4
105 cells were lysed in TeloChaser buffer (containing
2.3. Confocal microscopy – ds/siRNAs cellular uptake protease and phosphatase inhibitor cocktail, Toyobo Co. Ltd.) for
Cells (2
105 cells/ml) were transfected with fluorescein-labeled ds/ 30 min at 4 °C, centrifuged at 12 000
g/20 min/4 °C and the super-
siRNA (single dose 180 nM, Qiagene) using lipofectamine as described natant was dissolved 1:1 in 2
Laemmli sample buffer (1.1 M Tris–
above. Six hours transient transfection was applied and transfected HCl, pH 6.0, 3.3% SDS, 22% glycerol, 10% b-mercaptoethanol and
cells were cultured in RPMI-1640 medium. At different times of post- 0.001% bromophenol blue). Samples (in equal protein concentration)
transfection (1, 2, 4, 6 and 8 h), aliquots of the cells were sedimented by were heated at 95 °C for 10 min and were applied to 5% stacking,
centrifugation (1000
g/10 min), washed twice with PBS (Ca2þ and 4–12% resolving SDS–polyacrylamide gel. Electrophoresis was
Z. Zhelev et al. / FEBS Letters 570 (2004) 195–204 197

Fig. 1. (A) Sequences of chemically synthesized anti-bcr-abl ds/siRNAs and their sense mRNA/cDNA target sequences. TT indicates the deoxy-
thymidine dimer as 30 overhang. (B) Intracellular uptake of ds/siRNAs. Cells (2
105 cells/ml) was transfected during 6 h with fluorescein-labeled ds/
siRNA using lipofectamine as described in Section 2. The cells were sedimented by centrifugation, washed twice with PBS (Ca2þ and Mg2þ free), and
the fluorescein-labeled ds/siRNA, incorporated into the living cells (green light), were detected by fluorescent confocal microscope. (C) Scheme of
‘‘chronic treatment’’ of cells with ds/siRNA-mix or ds/siRNA-non-sense. ds/siRNA-mix consisted of three ds/siRNAs (3
60 nM).
198 Z. Zhelev et al. / FEBS Letters 570 (2004) 195–204

carried out in two steps: at 80 V for 15 min and 120 V for 2 h at RT. bioluminescence was used as a marker of cell viability. Briefly, the cells
BioRad Kaleidoskop protein standards were also applied for com- (siRNA/Glivec-treated or untreated) were suspended in the culture
parison. After electrophoresis, the separated protein fractions were medium to a concentration of 5
105 cells/ml and then were added in
transferred to a Hybond-P PVDF membrane (Amersham Bioscience), 90 ll aliquots to 96-well plates. CellTiter 96 kit (Promega) was added
using XCell II Blot Module (Novex). The transfer was carried out at in aliquots of 90 ll to each patch and incubated with cell suspensions
35 V for 18 h at 4 °C. The membranes were cut at 45 kDa (for b-actin) for 1 h, following the procedure recommended by the producer. The
and 210 kDa (for bcr-abl) molecular weight levels. A double antibody luminescence, produced by luciferase-catalyzed luciferin + ATP reac-
procedure was used to detect the proteins. The membranes were in- tion, was measured by MicroLumat LB96P (Perkin–Elmer). The data
cubated under agitation for 1 h at RT in a blocking solution (PBS, were normalized to the control group (consisted of untreated cells).
containing 5% dry skimed milk and 0.1% Tween 20), then at RT for 1 h Cell viability of siRNA/Glivec-treated cells was calculated as % of
in anti-c-abl (rabbit, Sigma; 1:100 dilution for bcr-abl detection) or luminescence in comparison with the control.
anti-b-actin monoclonal antibodies (mouse, Calbiochem; 1:20 000).
The antibody solution was removed and the membranes were washed 2.12. Microarray analysis
three times with PBS, containing 0.1% Tween 20. The membranes were The RT labeling and hybridization followed the protocol recom-
then incubated (1 h at RT under agitation) with horseradish peroxi- mended by Agilent Technologies Inc. intended for Agilent microarray
dase-conjugated goat anti-rabbit IgG (Sigma, 1:5000) for anti-c-abl analysis. Briefly, a 20-lg aliquot of each mRNA sample was reverse
and goat anti-mouse IgM (Sigma, 1:2000) for anti-b-actin antibody. transcribed into a cDNA probe with oligo(dT) primer and labeled
Chemiluminescence was detected immediately using ECL Advance nucleotides. The reaction was carried out in a solution containing
Western Blotting Detection Kit (Amersham Bioscience). 50 lM dATP/dGTP/dTTP, 25 lM dCTP, 25 lM cyanine 3 (Cy3)-
dCTP (for control non-treated sample) or cyanine 5 (Cy5)-dCTP (for
2.8. Flow cytometry ds/siRNA-mix- or Glivec-treated sample) (Enzo Diagnostics, Inc.) and
Anti-c-abl antibody (rabbit, Sigma) was labeled by ZenonAlexa 400 U MMLV reverse transcriptase at 42 °C for 1 h. The labeling
Fluor-488 Rabbit IgG Labeling Kit (Molecular Probes). Permeabili- reaction was terminated by incubating the reaction mixture at 70 °C
zation of cells (treated or non-treated with ds/siRNA or Glivec) for for 10 min. The RNA was then degraded by adding 0.05 lg RNaseIA,
Fluor-488-conjugated anti-c-abl antibody was carried out by IntraPrep followed by incubation at 37 °C for 30 min. Degraded RNA and un-
Permeabilization Reagent (Immunotech). The antibody–antigene in- incorporated nucleotides were removed using a QIAquick PCR Puri-
teraction was detected by flow cytometry (Beckman Coulter-Epics fication Kit (Qiagen Inc.) according to the instructions of Agilent
XL). Data were collected and analyzed by ‘‘XL System II software’’. Technologies Inc. Hybridization was carried out in 22 ll of a hy-
No cells were excluded from the analysis and 5000 cells were coun- bridization mixture containing cDNA probes, the labeled orientation
ted. The results were presented as a dot plot of Fluor-488 fluorescence marker (Deposition Control SP300; Operon Technologies Inc.) and
(side scatter, SS – y-axis in 0–128 scale, vs. Fluor-488-conjugated an- mouse Cot-1 DNA (Invitrogen Corporation) at 65 °C for 17 h. The
tibody, Fluor-488 – x-axis in 1–1000 scale) with quadrant markers glass slides were then washed with 0.5
SSC and 0.01% SDS at RT for
drawn to distinguish the cells, containing different levels of antibody– 5 min, and with 0.06
SSC at room temperature for 2 min. After
antigene complexes. immediately removing the wash buffer by centrifugation, the glass
slides were scanned using GenePix 4000B (Axon Instruments, Inc.)
2.9. Tyrosine kinase activity test containing a 532 nm laser for Cy3 measurement and a 635 nm laser for
The cells (treated or non-treated with ds/siRNA or Glivec) were Cy5 measurement. Scans were made with a pixel resolution of 5 lm, a
sedimented by centrifugation (1000
g/10 min), washed twice with PBS laser power of 100%, and a photomultiplier tube voltage of 600 V for
(Ca2þ and Mg2þ free) and re-suspended in lysis buffer (TeloChaser, the 532 nm laser and 520 V for the 635 nm laser.
Toyobo Co., Ltd.). The lysis was carried out at 4 °C for 30 min in the To eliminate the effect of lipofectamine on the gene profile of K-562
presence of orthovanadate. The cytosolic fraction, containing bcr-abl, cells, in siRNA protocol control cells consist of cells treated with
was obtained after centrifugation at 12 000
g/10 min, 4 °C. Bcr-abl lipofectamine only on the scheme, shown in Fig. 1C.
oncoprotein was isolated together with c-abl (p150) by immunopre- Normalization and analysis of microarray data. Sixteen-bit TIFF
cipitation, using anti-c-abl antibody (Sigma, 3 lg/ml) and protein A/G images produced by the Axon scanner were analyzed using the
agarose beads. The tyrosine kinase activity of immunoprecipitated en- GenePixPro 3.0 (Axon Instruments, Inc.) software package. After
zymes was analyzed, using CHEMICON’s PTK Assay kit (with slight obtaining Cy3 and Cy5 grayscale images, each pseudo-color image was
modifications). In briefly, a synthetic biotinylated poly[Glu:Tyr], 4:1 overlaid, and all spots in the ratio image were defined by accessing the
(2.5 lg/ml), containing multiple tyrosine residues, was used as a PTK gene list file that described the location of each gene on the microarray.
substrate. The PTK reaction was started in cell lysate by 1 mM ATP/10 The average of the signal intensity was subtracted from the median of
mM MgCl2 and continued 60 min at 37 °C. After quenching the enzyme background intensity and outputted with the UniGene and GenBank
reaction with an inhibitor (100 mM EDTA), both the phosphorylated descriptors to a Microsoft Excel data spreadsheet. Relative expression
and dephosphorylated substrates were immobilized to the streptavidin- levels were calculated by global normalization between two samples
coated 96-well plates. The fraction of phosphorylated substrate was using all detected genes, after the exclusion of spots annotated as
visualized spectrophotometrically at 450 nm, using HRP-conjugated ‘‘Agilent QC’’, ‘‘Agilent Blank’’, and ‘‘Buffer’’.
mouse anti-phosphotyrosine antibody (clone PY20) and an ensuing
chromagenic substrate reaction. The quantity of phosphate, incorpo- 2.13. Statistical analysis
rated into the tyrosine kinase substrate, was determined utilizing the One-way analysis of variance was employed, followed by Bonfer-
phosphopeptide standard curve. roni’s test for truly significant differences. Statistical significance was
defined at P < 0:05. The statistical procedures were performed with
GraphPad InStat software, version 2.04, USA. Data are expressed as
2.10. Cell proliferation test
mean  S.D.
Cells (2
105 cells/ml) were cultured in the presence or in the ab-
sence of ds/siRNAs or Glivec, as described above. The cell growth was
detected spectrophotometrically, using CellTiter AQ Proliferation
Assay Kit (Promega). The method is based on the detection of the 3. Results and discussion
number of viable cells in proliferation. A tetrazolium compound
(MTS) is bioreduced by cells into a formazan product in the presence The sequences of the originally designed and chemically
of an electron coupling reagent (phenazine methosulfatef), based on
the method of Mosmann [21]. The absorbance of formazan product at synthesized ds/siRNAs are shown in Fig. 1A. The DNA target
490 nm (OD490 nm ) was recorded using 96-well plate reader Inter- sequences were localized in the abl-part of bcr-abl major
MedImmunomini NJ-2300 (InterMed, Japan). breakpoint of Philadelphia chromosome, which is character-
ized by a bcr exon 3/abl exon 2 (b3a2) junction (an unique
2.11. Cell viability assay of normal lymphocytes sequence in K-562 cells). The selection of DNA target se-
The viability of siRNA- and Glivec-treated normal lymphocytes was quences aimed to avoid the most sensitive for point mutations
measured by CellTiter-GloTM luminescent cell viability assay. ATP area in abl-part of bcr-abl gene (around Tyr315 ), responsible for
Z. Zhelev et al. / FEBS Letters 570 (2004) 195–204 199

Fig. 2. (A) Effect of ds/siRNA-mix on the bcr-abl mRNA in K-562 cells. The cells were transfected with ds/siRNA-mix (3
60 nM) or ds/siRNA-
non-sense (180 nM) as described in Section 2. Cell lysates were obtained after 6 days treatment and mRNA was isolated. The synthesis of bcr-abl
cDNA was carried out by RT-PCR and the product was analyzed by TBA-electrophoresis on 1% agarose precast gel with ethidium bromide. Blots
from one typical experiment are shown in the figure. In the histograms, the results are expressed as bcr-abl cDNA/b-actin cDNA and were given as a
percentage from the control group. The value of bcr-abl cDNA/b-actin cDNA in control (cells treated with lipofectamine only) was considered as
100%. The data are means  S.D. from six independent experiments. *** P < 0:001 vs. control group. (B) Effect of ds/siRNA-mix on the level of bcr-
abl oncoprotein in K-562 cells – Western blot analysis. The cells were treated with ds/siRNA-mix (3
60 nM) or ds/siRNA-non-sense (180 nM) and
cell lysates were obtained after 6 days treatment as described in Section 2. The samples (in equal protein concentration) were dissolved 1:1 in 2

Laemmli buffer, heated at 95 °C for 10 min and subjected to SDS–polyacrylamide gel electrophoresis (80 V, 15 min; 120 V, 2 h at RT). The separated
protein fractions were transferred to a Hybond-P PVDF membrane overnight at 35 V (4 °C). The membranes were cut at 45 kDa (for b-actin) and
210 kDa (for bcr-abl) and were incubated as follows: 1 h at RT in a blocking solution, 1 h in anti-c-abl antibody (rabbit, Sigma, 1:100 for detection of
bcr-abl) or in anti-b-actin antibody (mouse, Calbiochem, 1:20 000), 1 h with HRP-conjugated goat anti-rabbit IgG (Sigma, 1:5000) for anti-c-abl or
goat anti-mouse IgM (Sigma, 1:2000) for anti-b-actin. Chemiluminescence was detected immediately using ECL Advance Western Blotting Detection
kit (Amersham Bioscience). Blots from one typical experiment are shown in the figure. Control blots are representative for cells treated with
lipofectamine only. In the histograms, the results are expressed as a percentage from the control group. The value of p210 bcr-abl protein in control
(cells treated with lipofectamine only) was considered as 100%. The data are means  S.D. from five independent experiments. ** P < 0:01 vs control
group. (C) Effect of ds/siRNA-mix on the protein tyrosine kinase activity in K-562 cells. The cells were treated with ds/siRNA-mix (3
60 nM) or ds/
siRNA-non-sense (180 nM) as described in Section 2 and cell lysates were obtained after 6 days treatment. Bcr-abl p210 oncoprotein was isolated
together with c-abl p150 by immunoprecipitation, using anti-c-abl antibody (Sigma, 3 lg/ml) and protein A/G agarose beads. The tyrosine kinase
activity of immunoprecipitated enzymes was analyzed, using CHEMICON’s PTK Assay kit. The fraction of phosphorylated substrate was visualized
spectrophotometrically at 450 nm, using HRP-conjugated anti-phosphotyrosine antibody (clone PY20) and ensuing chromagenic substrate reaction.
The data are presented as means  S.D. from five independent experiments. OD450 nm ¼ 1.0 corresponds to 20 ng P-Tyr-peptide substrate. The
calibration curve was linear in the range 10–40 ng P-Tyr-peptide substrate. Control groups are representative for cells treated with lipofectamine
only. ** P < 0:01 vs. control group. (D) Effect of ds/siRNA-mix on the cell proliferation activity of K-562 cells. The cells were cultured in the
presence or in the absence of ds/siRNA-mix (3
60 nM) or ds/siRNA-non-sense (180 nM) as described in Section 2. The cell proliferation activity
was detected spectrophotometrically, based on the reduction of MTS into a formazan product (absorbance at 490 nm). The data were calculated as a
percentage from control (cells, treated with lipofectamine only). The cell proliferating activity of control cells was considered as 100%. The results are
presented as means  S.D. from eight independent experiments. ** P < 0:01 vs. control group.
200 Z. Zhelev et al. / FEBS Letters 570 (2004) 195–204

bcr-abl amplification and development of drug-resistance dur- minimize the side-effects of lipofectamine during repetitive
ing long-term treatment with conventional antileukemia drugs transfections and it was established that the selected scheme
(as it has been described in the clinical trials with Glivec [10,11]) overcame this restriction. No significant changes in parameters
(Fig. 1A). Thus, the designed ds/siRNAs have a potential to analyzed were detected in K-562 cells, treated with lipofecta-
express their activity if they are applied in combination with mine only as shown in Fig. 1C.
Glivec during long-term treatment of bcr-abl-positive cells. The capacity of ds/siRNA-mix to reduce the amount of bcr-
Several commercially available liposomal transfection re- abl mRNA in K-562 cells was estimated by RT-PCR (Fig. 2A).
agents (Lipofectamine 2000 from Invitrogen, Superfect and The treatment of cells with ds/siRNA-mix markedly reduced
Lipofect from Qiagen) were tested for their ability to deliver (82%) the level of bcr-abl mRNA. Similar effect was obtained
ds/siRNAs into bcr-abl-positive K-562 cells, as well as to on the level of p210 bcr-abl oncoprotein, determined by im-
preserve ds/siRNAs against degradation in RPMI-1640 me- munoblot analysis (64% reduction was detected, Fig. 2B). ds/
dium. K-562 cells were found to be suitable for analysis be- siRNAs-treatment of K-562 cells provoked also a significant
cause they contained a functional RNA interference decrease of their bcr-abl/c-abl tyrosine kinase activity (57%,
mechanism to bind to siRNAs and to mediate mRNA degra- Fig. 2C) and cell proliferation capacity (50%, Fig. 2D).
dation, which was confirmed in preliminary experiments by Mismatch-ds/siRNA did not influence the parameters ana-
lamin A/C system (RNAi Starter kit, Qiagen, and anti-lamin lyzed, indicating the sequence specificity of ds/siRNA-mix.
A/C antibody – Chemicon). Transfection efficacy was also Obviously, the ‘‘chronic treatment’’ can overcome the tran-
analyzed using lamin A/C system. The preference was given to sient nature of RNAi and to guarantee comparatively strong
lipofectamine because of most optimal conditions for com- suppression of bcr-abl mRNA and oncoprotein expression.
paratively high cellular uptake (transfection efficacy reached Since the selected ds/siRNAs have a potential to target also
approximately 70% in K-562 cells, data not shown) and long c-abl mRNA (Fig. 1A) and to influence the homeostasis of
half-life of ds/siRNA sequences during transfection. The half- normal cells, we clarified the effect of ds/siRNA-mix
life of fluorescein-labeled ds/siRNA–lipofectamine complexes (3
60 nM) on the viability of normal lymphocytes after
in cell culture medium was 10 h, determined by HPLC. The ‘‘chronic treatment’’. It was observed that ds/siRNA-mix did
selection of ds/siRNA/lipofectamine ratio meant to avoid also not express cytotoxicity to normal lymphocytes, at least during
side-effects of lipofectamine on cell viability and proliferation. 6 days treatment on the scheme, as shown in Fig. 1C. This
No significant changes were detected in the viability and pro- result does not exclude an expression of cytotoxicity of the
liferation activity of leukemia cells, treated during 6 h with designed ds/siRNAs to normal cells during long-term treat-
lipofectamine alone or in combination with ds/siRNAs (data ment and the sequences targeted only to fusion part of bcr-abl
are not shown). The cells, treated with lipofectamine only, mRNA have a privilege.
were used as controls. It was established that only 20% of ds/ The efficiency of the designed ds/siRNAs was compared with
siRNAs degradated during 6 h incubation in the medium if the that of Glivec. Glivec was applied in a dose of 180 nM every two
siRNA sequences are in a complex with lipofectamine. days (as shown in Fig. 1C). To guarantee an equal effect of ds/
The time-dependent cellular uptake of ds/siRNAs was de- siRNA-mix and Glivec on the cell proliferation activity of K-
tected by fluorescent confocal microscopy, using fluorescein- 562 cells, the treatment with Glivec was carried out every two
labeled ds/siRNA (Fig. 1B, green light). A comparatively good days and the effects on the parameters analyzed were detected on
intracellular uptake of ds/siRNA was detected on the 6th hour the 8th day. The results in Fig. 3 demonstrate that Glivec de-
of the beginning of transfection. In contrast, ds/siRNA was creased markedly protein tyrosine kinase activity (73%) and
not detected in the cells without any lipofectamine (no fluo- cell proliferation capacity (54%), and provoked a compara-
rescence was observed in the cells). Based on the preliminary tively poor decrease of bcr-abl protein (14%).
experiments, 6 h transient transfection of ds/siRNAs using Recently, we demonstrated that relatively long-term
lipofectamine was applied and the transfected cells were re- ‘‘chronic’’ application (6 days and more) of anti-bcr-abl oligo-
placed in a new culture medium. DNAs markedly reduced the level of bcr-abl protein, but their
Bcr-abl protein is considered as ‘‘a protein – relatively stable sequence-specific effect was accompanied with influence of
to the antisense attack’’ and it is difficult to influence its level telomeric-associated proteins, increased telomerase activity
by single and short-term application of anti-bcr-abl sub- and restoration of cell proliferation capacity in the treated cells
stances. In our previous work, we demonstrated that relatively [8,9]. Therefore, the comparatively long-term siRNA interfer-
long-term ‘‘chronic’’ application (6 days and more) of anti-bcr- ence of bcr-abl oncogene has also a potential to influence
abl oligo-DNAs markedly reduced the level of bcr-abl proteins positively the expression of other oncogenes, as well as factors,
and influenced significantly the proliferation activity of K-562 responsible for the proliferation and immortalization of bcr-
[8,9]. In this context, anti-bcr-abl ds/siRNAs have also a po- abl-positive cells.
tential to manifest cytotoxicity and to suppress the prolifera- Similar effects have also been reported for Glivec. Brum-
tion during long-term treatment, as it has been assumed by mendorf et al. [22] found an extension of the telomeres in
other authors [5,6]. To guarantee enough time for induction of CML-patients after long-term treatment with Glivec in both
mRNA cleavage as well as for re-synthesis of the target on- chronic phase and blast crisis. Analyzing 517 samples from 206
coproteins, ‘‘chronic treatment’’ of K-562 cells with ds/siRNAs patients, the authors have been established that telomere
(mix of the three sequences in equal concentrations – 3
60 length from start of treatment up to day 144 is significantly
nM) was applied. The scheme of ‘‘chronic treatment’’ is shown shorter compared with patients treated for more than 144
in Fig. 1C. It includes three repetitive transfections of ds/siR- days. In patients with repeated measurements, a significant
NA-mix (every two days) and the effects on bcr-abl mRNA, increase in telomere length under treatment is observed. Me-
p210 and all other parameters were analyzed on the 6th day. dian telomere length in major remission has been found to be
The major problem in this experimental protocol was to significantly longer in comparison with patients without
Z. Zhelev et al. / FEBS Letters 570 (2004) 195–204 201

Fig. 3. (A) Effect of Glivec on the level of bcr-abl oncoprotein in K-562 cells – Western blot analysis. The cells were treated with Glivec and cell
lysates were obtained after 8 days treatment as described in Section 2. The samples (in equal protein concentration) were dissolved 1:1 in 2
Laemmli
buffer, heated at 95 °C for 10 min and subjected to SDS–polyacrylamide gel electrophoresis (80 V, 15 min; 120 V, 2 h at RT). The separated protein
fractions were transferred to a Hybond-P PVDF membrane overnight at 35 V (4 °C). The membranes were cut at 45 kDa (for b-actin) and 210 kDa
(for bcr-abl) and were incubated as follows: 1 h at RT in a blocking solution, 1 h in anti-c-abl antibody (rabbit, Sigma, 1:100 for detection of bcr-abl)
or in anti-b-actin antibody (mouse, Calbiochem, 1:20 000), 1 h with HRP-conjugated goat anti-rabbit IgG (Sigma, 1:5000) for anti-c-abl or goat anti-
mouse IgM (Sigma, 1:2000) for anti-b-actin. Chemiluminescence was detected immediately using ECL Advance Western Blotting Detection kit
(Amersham Bioscience). Blots from one typical experiment for each protein are shown in the figure. Control blots are representative for untreated
cells. In the histograms, the results are expressed as a percentage from the control group. The value of p210 bcr-abl protein in control was considered
as 100%. The data are means  S.D. from four independent experiments. * P < 0:05 vs. control group. (B) Effect of Glivec on the level of bcr-abl
oncoprotein in K-562 cells – flow cytometric analysis. The cells were treated with Glivec as described in Section 2. After 8 days treatment, the cells
were analyzed for the level of bcr-abl oncoprotein. Anti-c-abl antibody (rabbit, Sigma) was labeled by ZenonAlexa Fluor-488 Rabbit IgG Labeling
kit (Molecular Probes). Permeabilization of cells for the Fluor-488-conjugated anti-c-abl was carried out by IntraPrep Permeabilization Reagent
(Immunothech). The antibody–antigene interaction was detected by flow cytometry. Data are presented as a dot plot of Fluor-488 fluorescence (side
scatter, SS – y-axis, vs. Fluor-488-conjugated antibody – x-axis in 1–1000 scale) with quadrant markers drawn to distinguish the cells, containing
different levels of antibody–antigene complexes. In the histograms, quadrant H corresponds to the cells, containing maximum level of bcr-abl
(positive control), quadrant G corresponds to the cells without fluorescent marker and therefore without bcr-abl–antibody complexes (spontaneous
cell fluorescence, negative control), and the section between H and G (mentioned as E) corresponds to the cells, containing moderate or low level of
Fluor-488-conjugated c-abl antibody and therefore expressing moderate or low levels of bcr-abl oncoprotein in comparison with the positive control.
Histograms from one typical experiment are shown in the figure. Positive control is representative for Glivec-untreated cells. (C) Effect of Glivec on
the protein tyrosine kinase activity in K-562 cells. The cells were treated with Glivec as described in Section 2 and cell lysates were obtained after 8
days treatment. Bcr-abl oncoprotein was isolated together with c-abl by immunoprecipitation, using anti-c-abl antibody (Sigma, 3 lg/ml) and protein
A/G agarose beads. The tyrosine kinase activity of immunoprecipitated enzymes was analyzed, using CHEMICON’s PTK Assay kit. The fraction of
phosphorylated substrate was visualized spectrophotometrically at 450 nm, using HRP-conjugated anti-phosphotyrosine antibody (clone PY20) and
ensuing chromagenic substrate reaction. The data are presented as means  S.D. from six independent experiments. OD450 nm ¼ 1.0 corresponds to
20 ng P-Tyr-peptide substrate. The calibration curve was linear in the range 10–40 ng P-Tyr-peptide substrate. *** P < 0:001 vs. control group. (D)
Effect of Glivec on the cell proliferation activity of K-562 cells. The cells were cultured in the presence or in the absence of Glivec as it was described
in Section 2. The cell proliferation activity was detected spectrophotometrically after 8 days treatment, based on the reduction of MTS into a
formazan product (absorbance at 490 nm). The data were calculated as a percentage from control non-treated cells. The cell proliferating activity of
control cells was considered as 100%. The results are presented as means  S.D. from eight independent experiments. ** P < 0:01 vs. control group.
202 Z. Zhelev et al. / FEBS Letters 570 (2004) 195–204

response to treatment, measured either by cytogenetics, inter- mia cells. To verify the factors potentially responsible for side-
phase FISH, or quantitative RT-PCR. effects of Glivec and bcr-abl RNAi during long-term treatment,
Obviously, both pathways of bcr-abl suppression – siRNA- we provided a microarray analysis of gene profile in K-562 cells,
mediated cleavage of bcr-abl mRNA and Glivec-mediated ‘‘chronically’’ treated with ds/siRNA-mix or Glivec.
inhibition of already synthesized fusion protein, are in a cross- The expression of 162 and 104 genes was found to
talk with other oncogenes, apoptotic/antiapoptotic and cell decrease and increase (>3 times), respectively, in ds/siRNA-
proliferation factors. Both mechanisms keep a potential for treated cells versus 94 and 76 genes in Glivec-treated. The
restoration of the proliferation and immortalization of leuke- data in Tables 1 and 2 summarize the significant changes in

Table 1
Transcripts of interest suppressed >3 times in siRNA- or Glivec-treated K-562 cells – microarray analysis (P < 0:01)
Accession number Description siRNA-treated/ Glivec-treated/
non-treated non-treated
Factors, responsible for cell proliferation and immortalization
U47077 Human DNA-dependent protein kinase catalytic subunit – 5.359
M23102 Human trk proto-oncogene insert of pLM6 4.941 –
U51869 Human proto-oncogene Bcd orf1 and orf2 mRNA 4.461 3.115
AF152961 Human chromatin-specific transcription elongation factor FACT
140 kDa subunit mRNA 4.371 3.903
AJ000052 Human gene encoding splicing factor SF1, exons 2–8 3.982 3.234
D29767 Human mRNA for Tec protein tyrosine kinase 3.859 3.594
X07109 Human mRNA for protein-kinase C-beta II – 3.634
D51465 Nuclear factor I/X (CCAAT-binding transcription factor) 3.545 –
NM_00325 Tight junction protein 1 3.343 –
AF017789 Human putative transcription factor CA150 mRNA – 3.336
T36282 Guanylate kinase – 3.334
M54968 Human k-ras oncogene protein mRNA – 3.301
L23320 Human replication factor C large subunit mRNA 3.310 –
S49592 E2F transcription factor 3.255 –
M34057 Latent transforming growth factor beta binding protein 1 3.232 –
AU118354 ATP-binding cassette, sub-family G (WHITE), member 2 3.188 –
M32721 Hyman poly(ADP-ribose)-polymerase mRNA – 3.150
M64347 Human novel growth factor receptor mRNA 3.046 –
AB005659 ATP-binding cassette, sub-family C (CFTR/MRP), member 5 3.015 –
Factors, responsible for apoptosis induction
AF249273 Human Bcl-2-associated transcription factor 4.501 3.044
BI545796 Protein phosphatase 2, regulatory subunit B – 4.453
Y13247 Protein phosphatase 1, regulatory subunit 10 – 4.111
U03106 Human wild-type p53 activated fragment-1 4.310 –
AB002331 Death-associated transcription factor 1 3.241 –
NM_03334 Caspase-7, apoptosis-related cysteine protease 3.109 –

Table 2
Transcripts of interest overexpressed >3 times in siRNA- or Glivec-treated K-562 cells – microarray analysis (p < 0:01)
Accession number Description siRNA-treated/ Glivec-treated/
non-treated non-treated
Factors, responsible for cell proliferation and immortalization
M24779 Human protein kinase-related oncogene (PIM1) mRNA – 9.913
AA232339 Pim-1 oncogene 4.889 –
AA863383 Pim-2 oncogene – 3.074
BM020481 Tumor protein D52-like 1 4.673 –
J04111 Human c-jun proto oncogene, clone hCJ-1 4.323 4.490
X52125 Human alternatively spliced c-myb mRNA, clone pMbm-1 – 4.496
X52479/AF035594 Protein-kinase C-alpha mRNA for PK-C-alpha 4.171 3.134
AI498125 Pvt-1 oncogene homolog, MYC activator 4.104 3.881
U80017 Human basic transcription factor 2 p44 (MAPK activator) – 3.565
BG387620 v-myb myeloblastosis viral oncogene homolog 3.452 –
M96956 Teratocarcinoma-derived growth factor, clone CR-3 3.404 –
AF053949 Human transcription factor CBFA1/OSF2 gene 3.046 –
X52125 Human alternatively spliced c-myb mRNA – 4.496
Factors, responsible for apoptosis induction
M54894 Interleukin-6 (interferon, beta 2) 12.38 5.915
X57351 Human 1-8D gene from interferon-inducible gene family 5.098 4.707
AA769631 Tumor necrosis factor receptor superfamily, 10b 4.954 4.429
AF090693 Human apoptosis-related RNA binding protein 4.546 –
J04164 Human interferon-inducible protein 9-27 – 3.590
BC010399 STAT induced STAT inhibitor-2 3.402 3.780
NM00218 Interleukin-13 – 3.341
Z. Zhelev et al. / FEBS Letters 570 (2004) 195–204 203

genes of interest – kinases, apoptotic/antiapoptotic factors, treatment. It did not exclude many other genes to be influenced
cell proliferating factors and oncogenes. Table 1 includes the significantly during treatment more than 6–8 days as in the
transcripts with decreased expression in ds/siRNA- or Gli- present study. In this context, the combined application of
vec-treated K-562 cells (>3 times). RNAi of bcr-abl was Glivec (or anti-bcr-abl siRNAs) with other substances directed
accompanied with decreased expression of several proto- to the genes described above is a promising strategy for CML
oncogenes (trk and Bcd orf1/orf2 proto-oncogenes), growth control.
factors (novel growth factor receptor, latent transforming
growth factor b binding protein 1), factors relating to kinase
Acknowledgements: This study was supported in part by the JSPS
activity (Tec protein tyrosine kinase, ATP-binding cassette – Invitation Fellowship Program for Research in Japan. The GenBank
sub-families C-CFTR/MRP and G-WHITE) and factors re- Accession Nos. for human bcr-abl and human c-abl cDNA se-
lating directly to cell proliferation cycle (Table 1). Several quences described in this paper are, respectively, AJ131466.1 and
factors responsible for development of apoptosis also in- X16416.1. ds/siRNA constructs correspond to the nucleotide sites,
localized in the breakpoint cluster region (b3a2 junction) in the 9:22
creased significantly (>3 times): STAT induced STAT
chromosomal translocation (Philadelphia chromosome) – abl-gene
inhibitor-2, apoptosis-related RNA binding protein (NA- (Accession No. AJ131466 – between ex. 374  547 for ds/siRNA-2
POR-3), tumor necrosis factor receptor superfamily (member and ex. 548  843 for ds/siRNA-1 and ds/siRNA-3). ds/siRNA-4 is a
10b), interleukin 6 (Table 2). non-sense sequence and did not show homology with cDNAs of c-
However, 6 days treatment of K-562 cells with anti-bcr-abl abl as well as bcr-abl gene.
ds/siRNA-mix resulted in a significant decrease of several ap-
optotic factors (caspase-7, p53, death associated transcription
factor 1) and antiapoptotic factors (bcl-2-associated tran- References
scription factor) (Table 1), as well as in a significant increased
expression of the following transcripts that can be responsible [1] Van Etten, R.A. (2002) Studying the pathogenesis of Bcr-Abl+
for restoration of the cell proliferation (Table 2): protein ki- leukemia in mice. Oncogene 21, 8643–8651.
[2] O’Dwyer, M.E., Mauro, M.J. and Druker, B.J. (2002) Recent
nase-related oncogene (PIM1), c-jun proto-oncogene, Pvt-1 advancements in the treatment of chronic myelogenous leukemia.
oncogene homologue (myc-activator), protein kinase C-a gene, Annu. Rev. Med. 53, 369–381.
teratocarcinoma-derived growth factor 3, transcription factor [3] Wisniewski, D., Lambek, C.L., Liu, C., Strife, A. and Veach,
CBFA1/OSF2. D.R., et al. (2002) Characterization of potent inhibitors of the Bcr-
Abl and the c-kit receptor tyrosine kinase. Cancer Res. 62, 4244–
Eight days treatment of K-562 cells with Glivec was ac-
4255.
companied with suppression of Bcd orf1/orf2 proto-oncogene, [4] Fabbro, D., Ruetz, S., Buchdunger, E., Cowan-Jacob, S.W. and
k-ras oncogene, factors relating to kinase activity (Tec protein Fendrich, G., et al. (2002) Protein kinases as targets for anticancer
tyrosine kinase, protein kinase C-b) and factors relating di- agents: from inhibitors to useful drugs. Pharmacol. Ther. 93, 78–
rectly to cell proliferation cycle (Table 1). Glivec also sup- 98.
[5] Wilda, M., Fuchs, U., Wossmann, W. and Borkhardt, A. (2002)
pressed protein phosphatase 1 and 2 genes and thus can Killing of leukemic cells with a bcr/abl fusion gene by RNA
stimulate indirectly the activity of other protein kinases. Sig- interference (RNAi). Oncogene 21, 5716–5724.
nificant overexpression of the following transcripts was [6] Scherr, M., Battmer, K., Winkler, T., Heidenreich, O., Gasner, A.
detected in Glivec-treated K-562 cells (Table 2): protein and Eder, M. (2003) Specific inhibition of bcr-abl gene expression
by small interfering RNA. Blood 101 (4), 1566–1569.
kinase-related oncogene (PIM2), c-jun proto-oncogene, Pvt-1
[7] Wohlbold, L., van der Kuip, H., Mieting, C., Vornlocher,
oncogene homologue (myc-activator), protein kinase C-a gene, H.-P., Knabbe, C., Duyster, J. and Aulitzky, W.E. (2003)
human basic transcription factor 2 p44 (MAPK activator). Inhibition of bcr-abl gene expression by small interfering RNA
However, Glivec induced an overexpression of several apop- sensitizes for imatinib mesylate (STI571). Blood 102 (6), 2236–
totic factors as: interleukins 6 and 13, 1-8D gene from inter- 2239.
[8] Bakalova, R., Ohba, H., Zhelev, Z., Ishikawa, M., Shinohara, Y.
feron-inducible gene family, tumor necrosis factor receptor and Baba, Y. (2003) Cross-talk between BCR-ABL tyrosine
superfamily (10b), interferon-inducible protein 9-27, STAT- kinase, protein kinase C and telomerase – a potential reason for
induced STAT-inhibitor 2. resistance to Glivec in chronic myelogenous leukemia. Biochem.
In summary, the results from microarray analysis dem- Pharmacol. 66 (10), 1879–1884.
[9] Bakalova, R., Ohba, H., Zhelev, Z., Kubo, T., Fujii, M.,
onstrate that siRNA interference of bcr-abl expression and
Ishikawa, M., Shinohara, Y. and Baba, Y. (2004) Antisense
inhibition of already synthesized bcr-abl fusion protein by inhibition of bcr-abl/c-abl synthesis promotes telomerase activity
Glivec are in a cross-talk with several common factors that and upregulates tankyrase in human leukemia cells. FEBS Lett.
might be responsible in major degree for cell death or 564, 73–84.
for restoration of the cell proliferation after long-term [10] Von Bubnoff, N., Schneller, F., Peschel, S. and Duyster, J. (2002)
BCR-ABL gene mutations in relation to clinical resistance of
treatment. Philadelphia-chromosome-positive leukaemia to STI571: a pro-
Both, anti-bcr-abl RNAi and Glivec-treatment of K-562 spective study. Lancet 359, 487–491.
cells decreased in equal degree bcl-2-associated transcription [11] Hofmann, W.K., Jones, L.C., Lemp, N.A., De Vos, S.,
factor (bcl-2 is one of the major apoptotic signals), and over- Gschaidmeier, H., Hoelzer, D., Ottmann, O.G. and Koeffler,
expressed in equal degree c-jun proto-oncogene, protein kinase H.P. (2002) Ph (+) acute lymphoblastic leukemia resistant to the
tyrosine kinase inhibitor STI571 has a unique BCR-ABL gene
C-a, pvt-1 oncogene homologue (myc-activator) (all they are mutation. Blood 99 (5), 1860–1862.
well-known antiapoptotic factors). It can hypothesized that [12] Paddison, P.J., Caudy, A.A. and Hannon, G.J. (2002) Stable
bcr-abl oncoprotein participates in the regulation of these suppression of gene expression by RNAi in mammalian cells.
common genes and/or their respective proteins. To guarantee Proc. Natl. Acad. Sci. USA 99, 1443–1448.
[13] Elbashir, S.M., Harbort, J., Lendeckel, W., Yalcin, A., Weber, K.
the success of siRNA interference and Glivec, it is necessary to and Tuschl, T. (2001) Duplexes of 21-nucleotide RNAs mediate
clarify the expression of all these factors and to prognosticate RNA interference in cultured mammalian cells. Nature 411 (6836),
their role in restoration of cell proliferation during long-term 494–498.
204 Z. Zhelev et al. / FEBS Letters 570 (2004) 195–204

[14] Montgomery, M.K., Xu, S. and Fire, A. (1998) RNA as a target of [19] Ardelt, B., Ardelt, W. and Darzynkiewicz, Z. (2003) Cytotoxic
double-stranded RNA-mediated genetic interference in Caenor- ribonucleases and RNA interference (RNAi). Cell Cycle 2 (1), 22–
habditis elegans. Proc. Natl. Acad. Sci. USA 95, 15502–15507. 24.
[15] Baulcombe, D.C. (1999) Fast forward genetics based on virus- [20] Wall, N. (2003) Snall RNA: can RNA interference be exploited
induced gene silencing. Curr. Opin. Plant. Biol. 2, 109–113. for therapy. Lancet 362, 1401–1403.
[16] Kennerdell, J.R. and Carthew, R.W. (2000) Heritable gene [21] Mosmann, T. (1983) Rapid colorimetric assay for cellular growth
silencing in Drosophila using double-stranded RNA. Nat. Bio- and survival: application to proliferation and cytotoxicity assays.
technol. 18, 869–898. J. Immunol. Meth. 65, 55–63.
[17] Scherr, M., Morgan, M.A. and Eder, M. (2003) Gene silencing [22] Brummendorf, T.H., Ersoz, I., Hartmann, U., Balabanov,
mediated by small interfering RNAs in mammalian cells. Curr. S., Wolke, H., Paschka, P., Lahaye, T., Berner, B.,
Med. Chem. 10 (3), 245–256. Bartolovic, K., Kreil, S., Berger, U., Gschaidmeier, H.,
[18] Saunders, L.R. and Barber, G.N. (2003) The dsRNA binding Bokemeyer, C., Hehlmann, R., Dietz, K., Lansdorp, P.M.,
protein family: critical roles, diverse cellular functions. FASEB J. Kanz, L. and Hochhaus, A. (2003) Ann. N.Y. Acad. Sci.
17 (9), 961–983. 996, 26–38.