Welcome international healthcare professionals

This site is no longer supported and will not be updated with new content. You are welcome to browse and download all content already included in the site. Please note you will have to register your email address to access the site.

You are here

Imatinib resistance in multidrug-resistant K562 human leukemic cells

Leukemia Research, 5, 33, pages 710 - 716


The multidrug resistance phenotype (MDR) is one of the major causes of failure in cancer chemotherapy and it is associated with the over-expression of P-glycoprotein (P-gp or MDR1) in tumor cell membranes. A constitutive NF-κB activity has been observed in several haematological malignancies and this is associated with its anti-apoptotic role. In the present work, the relationship between NF-κB and MDR phenotype was evaluated in wild type K562 human leukemic cells (K562-WT) and in its vincristine-resistant counterpart, K562-Vinc cells. These data showed that K562-Vinc cells, which express an active P-gp, exhibited MDR phenotype. The resistant indexes (IC50K562-Vinc/IC50K562-WT) for structurally unrelated drugs like imatinib, doxorubicin and colchicine were 8.0 ± 0.3, 2.8 ± 0.4 and 44.8 ± 8.8, respectively. The imatinib resistance was reversed by P-gp blockade suggesting the involvement of P-gp in imatinib transport. We observed that NF-κB was constitutively activated in both cell lines but in a lesser extent in K562-Vinc. The inhibition of NF-κB with BAY 11-7082 increased the cytotoxicity of imatinib in K562-Vinc cells but not in K562-WT. Further, the co-administration of imatinib and BAY 11-7082 sensitized multidrug-resistant K562 cells to cell death as detected by increased percentage of annexin V positive cells. The induced cell death in K562-Vinc cells was associated with activation of caspases 9 and 3. Finally, we provide data showing that BAY 11-7082 down-regulates the expression of P-gp suggesting that the activity of NF-κB could be functionally associated to this protein in K562 cells. Our results indicate that the vincristine-resistant K562 cells which developed MDR phenotype, exhibited resistance to imatinib associated with a functional P-gp over-expression. This resistance could be partially overcome by the inhibition of NF-κB pathway.

Keywords: MDR1, P-glycoprotein, K562 cells, Imatinib, NF-κB, BAY 11-7082.

1. Introduction

The multidrug resistance phenotype (MDR) is considered one of the major causes of failure in cancer chemotherapy and it is associated with the over-expression of P-glycoprotein (P-gp or MDR1) in tumor cell membranes. P-gp, the product of the MDR1 gene, is an ATP-dependent pump that transports a wide variety of structurally diverse compounds out of the cell, resulting in a decrease of its intracellular accumulation (for reviews see [1], [2], and [3]) and it is among the strongest prognostic factors in acute myeloid leukemia [4] .

The molecular mechanisms associated to the expression control of P-gp are not complete elucidated, even though evidences suggest that NF-κB may be one of the transcription factors involved [5], [6], [7], and [8]. NF-κB is the prototype of a family of transcription factor made by dimerization of five subunits: p65 (RelA), c-Rel, RelB, p50, and p52. Although many dimeric forms of NF-κB have been detected, the more ubiquitous hetero-dimer is the p65 and p50. Before cell stimulation, most of NF-κB is present in the cytoplasm as an inactive complex consisting of Rel hetero-dimer and the inhibitor subunit (IκB). After stimulation, IκB undergoes phosphorylation and ubiquitination-dependent degradation by the proteosome, thus NF-κB dimers translocate to the nucleus where they bind to a specific consensus sequence in the DNA resulting in the activation of target genes transcription [9] . BAY 11-7082, is an inhibitor of IκBα phosphorylation which leads to reduced IκBα proteasomal degradation, thereby resulting in a decreased nuclear NF-κB [10] and [11]. NF-κB is a mediator of inducible gene expression in response to inflammatory cytokines, pathogens and several stress signals, and is known for its crucial roles in the immune system, cell proliferation and transformation, apoptosis and tumor development [12], [13], and [14]. A constitutive NF-κB activity has been observed in several haematological malignancies [15], [16], and [17].

In order to investigate if the inhibition of NF-κB may overcome drug resistance we used a human myeloid K562 cells resistant to vincristine K562-Vinc, and their sensitive counterparts K5652-WT [18] . These chronic myeloid leukemic (CML) cells are characterized by the presence of a bcr-abl fusion gene, which produce a chimeric protein known as bcr-abl with constitutive tyrosine kinase activity. Imatinib mesylate emerged as a very useful compound for clinical development, since it potently inhibits all of the ABL tyrosine kinases and it has proven to be highly effective in the treatment of the CML [19] and [20]. However, some leukemia relapse can occur and several mechanisms of resistance were well described [21], [22], and [23].

In this study we confirmed that vincristine-resistant cells exhibit the MDR phenotype by determining the cross-resistance to structurally unrelated drugs like doxorubicin, colchicine and imatinib. We also show that the resistance to imatinib exhibited in these multidrug-resistant human leukemic K562 cells was mediated by P-gp and reversed by blockade of NF-κB pathway using the specific inhibitor BAY 11-7082.

2. Materials and methods

2.1. Cell culture

The human chronic myeloid leukemia K562 cell line (American Type Culture Collection, USA) was maintained in RPMI 1640 culture medium (Hyclone, UT, USA) supplemented with 10% fetal bovine serum (Natocor, Argentina), at 37 °C in a humidified atmosphere of 5% CO2 in air. The vincristine-resistant K562 cells (K562-Vinc) obtained by subculturing K562 cell in stepwise increased concentration of vincristine were provided by Dr Lehne G. (University of Oslo, Norway). K562-Vinc cells were continuously grown in the presence of 150 nM vincristine sulphate (Filaxis, Argentina) as described before [24] .

2.2. Proliferation assay

Cell proliferation was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT), assay (Sigma, St. Louis, MO, USA). 100 μl cell suspension (3 × 105 cell/ml) was incubated in 96-well culture plates in RPMI medium without phenol red and in the presence or absence of the test drugs for the stipulated times. Incubation with vehicle alone was performed as control. After that, 20 μl of MTT (5 mg/ml in PBS) were added to each well and incubated for 4 h at 37 °C. The MTT formazan precipitate was dissolved in 100 μl of DMSO. The optical density was measured at 570 nm (OD570nm) on an ELISA plate reader (R&D Systems, Minneapolis, USA) and the cell proliferation for each treatment was calculated as a percentage of control: treated cultures OD570nm/control cultures OD570nm × 100 and expressed as mean ± standard deviation. A sigmoidal function was used to characterize the dose–response relationship for each drug when administered alone or combined.

2.3. Protein extracts preparation

To prepare K562-WT or K562-Vinc total extracts, cells were washed twice with cold phosphate-buffered saline (PBS), and resuspended in a small volume of lysis buffer (50 mM HEPES, 250 mM NaCl, 1 mM EDTA, 1% NP-40, 1 mM PMSF, 1 mM DTT and protease inhibitor cocktail, Sigma, St. Louis, MO). Nuclei and unbroken cells were removed with centrifugation (1500 rev min−1 for 15 min, at 4 °C). For nuclear extracts preparation, cells were washed twice with cold PBS, resuspended in 500 μl of hypotonic buffer containing 10 mM HEPES, 10 mM KCl, 1.5 mM MgCl2, 1 mM DTT and 0.5% NP-40 and centrifuged at 3000 rev min−1 for 5 min. The pellets were washed three times in hypotonic buffer without NP-40 and suspended in 50 μl of ice-cold lysis buffer (20 mM HEPES, 420 mM NaCl, 1.5 mM MgCl2, 25% glycerol, 0.2 mM EDTA, 1 mM PMSF, 1 mM DTT and protease inhibitor cocktail) and left on ice for 30 min. Extracted nuclei were isolated by further centrifugation at 12000 rev min−1 for 10 min at 4 °C. Total protein in each sample was quantified according to the Bradford method (Sigma, St. Louis, MO).

2.4. Western blot

The samples were boiled for 10 min in buffer (50 mM Tris, pH 6.8, 2% (w/v) SDS, 10% (v/v) glycerol, 100 mM DTT, and 0.2 mg/ml bromophenol blue). Thirty micrograms of total protein for each condition were separated by 6–12% SDS-polyacrylamide gels and electrotransferred onto nitrocellulose membranes (Hybond ECL, Amersham Pharmacia Biotech Ltd., UK). These membranes were blocked with 10% (w/v) non-fat dry milk in Tris-buffered saline with 0.05% Tween (T-TBS) for 1 h at room temperature, and then incubated with primary antibodies against MDR1, p65/RelA, procaspase 9, α-tubulin (Santa Cruz Biotechnology, Inc., CA, USA) or procaspase 3 (Transduction Lab, BD Bioscience). The membranes were washed with T-TBS and then incubated with goat anti-rabbit IgG antibody (for MDR1, p65 and procaspase 9) or horse anti-mouse IgG antibody (for α-tubulin and procaspase 3), both conjugated to horseradish peroxidase (Vector Laboratories, Burlingame, CA, USA). The bands were detected by chemiluminescence. The relative level of each protein was obtained using the ImageJ 1.37V densitometric software (NIH, USA).

2.5. Fluo-3/AM accumulation assay

K562-WT or K562-Vinc cells were resuspended in PBS (without calcium and magnesium) at a concentration of 2 × 106/ml, aliquoted in 200 μl and prewarmed for 10 min at 37 °C. Then, Fluo-3/AM (Sigma, St. Louis, MO, USA) was added directly to each tube (final concentration 4 μM) and the mixtures were incubated for additional 45 min. The samples were washed twice with PBS/azide and the cell-associated mean fluorescence (excitation wavelength = 488 nm; emission wavelength = 525 nm) was measured using a FACSCalibur cytometer (Becton–Dickinson). To evaluate the effect of MDR reversion on intracellular accumulation of Fluo-3, 10 μM verapamil (Sigma, St. Louis, MO, USA) was added just before the addition of fluorescent agent. Cells incubated without Fluo-3/AM but with verapamil were used in control experiments.

2.6. Electrophoretic mobility shift assay (EMSA)

The EMSA was performed as described previously [25] . Briefly, cells were stimulated by 2 h with 2.5, 10 or 20 μM of BAY 11-7082, and the nuclear extracts were obtained by using the lysis buffer A (HEPES 10 mM, pH 7.9, MgCl2 1.5 mM, KCl 10 mM, NP-40 0.5%). Nuclear proteins were prepared by washing the nuclear pellet with buffer B (HEPES 10 mM, pH 7.9, MgCl2 1.5 mM, KCl 0.42 M, EDTA 0.25 M, glycerol 20%) by continuous shaking for 15 min at 4 °C. Binding to DNA reactions was performed in buffer C (HEPES 10 mM, pH 7.9, KCl 60 mM, EDTA 0.25 M, glycerol 20%). All the buffers contained the protease inhibitors 10 mg/ml leupeptin, 10 mg/ml aprotinin, 1 mg/ml pepstatin A, 1 mM phenylmethylsulfonyl fluoride and 1 mM dithiothreitol. Five micrograms of nuclear extract proteins were preincubated with 200 ng of poly(dI-dC) in binding buffer C. Then, 1 ng of 32P end-labeled double-stranded κB oligonucleotide, containing the sequence 50-AGTTGAGGGGACTTTCCCAGGC-30 was added to each tube and incubated for 15 min at room temperature. For supershift experiments, samples were preincubated for 1 h with the specific antibodies. In all, 5–20-fold excess of the unlabeled κB oligonucleotide or the unrelated oligonucleotide CCAAT (50-GATCCCGGAGCCCGGGCCAATCGGCGCA-30) was utilized for competition assays. DNA-protein complexes were separated in non-denaturing 5% polyacrylamide gels. The gels were dried under vacuum and autoradiographed at −70 °C.

2.7. Measurement of apoptosis

Following treatment with imatinib and/or BAY 11-7082, K562-WT and K562-Vinc cells were analyzed for translocation of phosphatidylserine to the outer surface of the plasma membrane, which is a marker of apoptosis, using the human phospholipid-binding protein annexin V-FITC (BD Pharmingen). Briefly, the cells were washed, resuspended in 100 μl of 1× binding buffer and then, 5 μl of annexin V-FITC plus 10 μl of propidium iodide (Sigma, USA) were added. Samples were incubated at room temperature for 15 min, after that 400 μl of 1× binding buffer was added and the fluorescence was determined by flow cytometry (FACSCalibur, Becton–Dickinson). Annexin V-FITC positive cells were analyzed by quadrant statistics using WinMDI 2.8 software (Scripps Research Institute, USA).

2.8. Statistical analysis

Data were expressed as mean ± S.D. (n = number of cells analyzed). Sets of measurement data were compared for statistical significance using either a Student’s t-test or ANOVA. Differences were considered statistically significant when p < 0.05.

3. Results

3.1. Expression and functionality of P-gp

The expression of P-gp in K562 cells resistant to vincristine was evaluated by Western blot using specific antibody. A ∼170 kDa expected band was observed in K562-Vinc cells while this protein was not detected in the parental K562 cells, indicating an over-expression of P-gp in a resistant variant ( Fig. 1 A). Although the resistant K562 cells were characterized previously [26] , we considered important to confirm the maintenance of P-gp expression and its functionality in our culture conditions. In order to analyze the P-gp activity, we determined the P-gp mediated transport by measuring the intracellular accumulation of the fluorescent probe Fluo-3 by flow cytometry ( Fig. 1 B). The fluorescence intensity were 29.9 ± 1.0 and 9.1 ± 0.7 arbitrary unit, for K562-WT and K562-Vinc cells, respectively (p < 0.01, n = 3). The accumulation of Fluo-3 in K562-Vinc cells was strikingly increased in the presence of 10 μM verapamil (p < 0.01, n = 3). We observed a slight effect of this inhibitor in wild type cells on fluorescence probe accumulation, in spite of not detecting P-gp by immunoblot. Previously, we also detected the P-gp mRNA expression in both cell lines by semi-quantitative RT-PCR, where the K562-Vinc showed a 20-fold increase in the P-gp levels compared with K562-WT cells [24] . This variability could be influenced by differences in the sensitivity between the employed techniques; however our results clearly indicate that K562-Vinc cells over-express a functional P-gp protein.


Fig. 1 (A) Representative Western blot of total protein extracts from K562-WT and K562-Vinc cells. A band at 160–170 kDa was detected only in resistant cells using specific antibody against P-gp. α-Tubulin was used as internal control. (B) Histograms of Fluo-3 accumulation patterns in the presence (solid lines) or absence (dotted lines) of 10 μM verapamil for K562-WT y K562-Vinc. The data are representative of those obtained in three other experiments.

3.2. Determination of multidrug resistance phenotype

The cytotoxicity of structurally unrelated drugs was evaluated in parental K562-WT and K562-Vinc cells by MTT assay in order to determine if the vincristine-selected cells show a MDR phenotype. The relative cell proliferation was measured after the incubation with increasing concentrations of vincristine, doxorubicin, colchicine and imatinib for 72 h. We found that the drug concentrations that reduced cell proliferation by 50% (IC50) for all tested drugs were significantly greater in K562-Vinc than these found in the parental K562 cells (p < 0.01, n = 5). The resistance index (RI) calculated by dividing the IC50 of resistant cells by the IC50 of parental cells for each drug has been widely used to evaluate de resistance level [27] . Thus, the calculated RI for vincristine, doxorubicin, colchicine and imatinib in K562 cells were 505.4 ± 278.9, 10.3 ± 6.3, 44.8 ± 8.8 and 8.1 ± 2.8, respectively ( Table 1 ). Similar RI values were obtained by determining the viable cells after the drugs incubation using the Trypan blue dye exclusion method (data not shown). The doubling time and morphology were not different for the parental K562 and the resistant counterpart (doubling time 33.7 ± 4.8 and 35.2 ± 3.4 h, for K562-WT and K562-Vinc, respectively, n = 3). Thus the dissimilar IC50 values found between both cell types for each drug should not be due to a difference in a basal proliferation rate. The wild type and resistant cells showed comparable sensitivity to aclarubicin and hydrogen peroxide after 72 h of incubation, indicating that these drugs are non-P-gp substrates; the RI were 0.97 ± 0.55 and 1.15 ± 0.46, respectively (n = 3). In addition, both cell lines were equally sensitive to serum starvation.

Table 1 Cross-resistance to structurally unrelated drugs was evaluated in K562-WT and K562-Vinc a .

Drug IC50 RI b
  K5 62-WT K5 62-Vinc  
Vincristine 1.1 ± 1.4 nM 533.2 ± 294.2 nM 505.4 ± 278.9
Doxorubicin 65.0 ± 35.4 nM 62 5.0 ± 7.1 nM 10.3 ± 6.3
Colchicine 10.3 ± 1.5 nM 453.3 ± 49.3 nM 44.8 ± 8.8
Imatinib 0.3 ± 0.2 μM 2.2 ± 0.7 μM 8.1 ± 2.8

a Cytotoxic effects of vincristine, colchicine, doxorubicin and imatinib was evaluated by MTT assay after 72 h of incubation. The IC50 values for each drug are presented as mean ± S.D. from five independent experiments.

b RI was calculated by dividing the IC50 of K562-Vinc cells by the IC50 of K562-WT cells.

We evaluated the contribution of P-gp in the imatinib resistance observed in K562-Vinc cells since it was described amplification or mutations of the bcr-abl gene as the main mechanisms leading to imatinib resistance [21] and [22]. Cells were incubated with a P-gp inhibitor verapamil (10 μM), and assessed for viability when exposed to imatinib for 72 h. In this condition the RI changed from 7.0 ± 2.4 to 2.0 ± 0.9 (p < 0.01, n = 4) showing a clear enhancement of imatinib sensitivity ( Fig. 2 ). Further, although K562-Vinc cells have never been exposed to imatinib before these experiments, we determined the expression of bcr-abl to discard amplification of this gene as a mechanism leading to imatinib resistance. Real-time PCR analysis revealed that the amplification of bcr-abl gene is not greater in resistant cells compared to parental K562 cells (data not shown). Taken together all these results strongly demonstrate that P-gp over-expression in K562-Vinc cells could be associated to a MDR phenotype and moreover, it is also contributing to the imatinib resistance in these cells.


Fig. 2 The effect of P-gp inhibition over imatinib cytotoxicity was evaluated in K562-WT and K562-Vinc by MTT assay (see Section 2 ). The cell proliferation was analyzed at the indicated imatinib concentration in the absence or presence of 10 μM verapamil after 72 h of incubation. Data represent mean ± S.D. from four separate experiments.

3.3. Constitutive NF-κB activity in both K562 cell lines

Several reports described a constitutive NF-κB activity in several malignant haematological cells [15], [16], and [28]. We examined the activity of this transcription factor in both K562 cell lines. Electrophoretic mobility shift assays (EMSAs) were performed in order to measure the binding of the available NF-κB in nuclear extracts obtained from K562-WT and K562-Vinc cells to a 32P-labeled NF-κB consensus oligonucleotide as described in Methods. We observed that NF-κB DNA-binding was constitutively activated in both cell lines but in a lesser extent for resistant K562-Vinc cells. The same results were obtained in four independent experiments and a representative experiment is shown in Fig. 3 A. To assess whether the permanent exposure to vincristine in K562-Vinc cells would somehow down-regulates the NF-κB pathway, we performed EMSA in nuclear extracts obtained from cells cultured without the drug. The vincristine was removed from de culture medium 24 h before the experiments. We observed comparable activity of NF-κB in both conditions. The nature of the DNA-binding complexes was revealed in supershift experiments using anti-p50 or anti-p65 specific antibodies. These assays have demonstrated that the dimmer p65/p50 is the most abundant complexes in both cell lines. The specificity of these complexes was confirmed by competition assays using an excess of unlabeled consensus oligonucleotide. Furthermore, in order to explore if the less constitutive activity of NF-κB observed in a resistant cells was associated with a decreased total p65 expression (nuclear and cytosolic), we performed Western blot analysis using a specific antibody against p65 ( Fig. 3 B). We detected similar amounts of total p65 in K562-WT and K562-Vinc cell lysates, being the relationship p65/α-tubulin 1.0 ± 0.2 and 1.1 ± 0.7, respectively (p > 0.5, n = 3). Therefore, our results indicate that K562-Vinc cells have similar levels of NF-κB than the wild type cells, but show a lower constitutive activity.


Fig. 3 (A) Analyses of DNA-binding activity in K562 cells. Left: EMSA experiments were performed by incubating nuclear extracts from K562-WT and K562-Vinc cells with oligonucleotide probes containing the consensus κB site. The (#) indicates that vincristine was removed from de culture medium 24 h before the nuclear extracts obtaining. Middle: Supershif experiments were carried out to identify the components of NF-κB complexes in both types of cells. The specific antibodies to p65 and p50 subunits used to supershif were indicated at the top. Right: For competition assays, nuclear extracts of K562-WT were preincubated with 5-, 10- and 20-fold molar excess of unlabeled NF-κB wild type oligonucleotides prior to addition of labelled NF-κB oligonucleotides. Lane C (control) was loaded with labelled oligonucleotides. The non-specific bands were identified as n.s. All these results are representative of at least three independent experiments. (B) Expression of total (nuclear and cytoplasmic) p65/RelA was analyzed by Western blot in total lysates from K562-WT and K562-Vinc cells. The relative level of p65 in each cell line using α-tubulin as internal control was represented in the bar graph (p > 0.05, n = 3).

3.4. Effect of BAY 11-7082 over imatinib cytotoxicity

In view of our results, we then analyzed the effect of NF-κB activity over the imatinib sensitivity. We determined whether incubation with the NF-κB inhibitor, BAY 11-7082, was associated with changes in the imatinib induced cytotoxicity. Fig. 4 A shows the relative K562-WT and K562-Vinc cell proliferation determined by culturing these cells with increasing imatinib concentrations with or without 2.5 μM BAY 11-7082 for 72 h. The sensitivity of imatinib did not change in the presence of NF-κB inhibitor for K562-WT cells. The IC50 were 0.26 ± 0.17 and 0.18 ± 0.15 μM in the absence and presence of BAY 11-7082, respectively (p > 0.05, n = 5). In contrast, co-administration of BAY 11-7082 significantly increased the sensitivity to imatinib in K562-Vinc cells. The IC50 were 2.24 ± 0.93 and 0.95 ± 0.65 μM in the absence and presence of the inhibitor, respectively (p < 0.05, n = 5). We also found an additional increase in the cytotoxicity induced by imatinib when BAY 11-7082 and verapamil were combined. We observed from this experiments that 10 μM verapamil further reduced the cell proliferation by 20% (18.9 ± 12.7% at 1 μM of imatinib, n = 6). The values of IC50 for imatinib in wild type and MDR K562 cells were similar to those published by Mukai et al. [29] .


Fig. 4 (A) The effect of 2.5 μM BAY 11-7082 over imatinib-induced cytotoxicity was evaluated in K562-WT and K562-Vinc by MTT assay. Bars represent the cell proliferation analyzed at the indicated concentrations of drugs after 72 h. The asterisk indicates statistical significance between bars (n = 5). (B) EMSA experiments were performed to analyze the DNA-binding activity of NF-κB in K562-WT and K562-Vinc cells after BAY 11-7082 incubation for 24 h at the indicated concentrations. Three additional experiments yielded similar results.

On the other hand, we performed EMSA to test whether the concentration of BAY 11-7082 that we used in a previous assay modifies the activity of NF-κB in both cell lines. Representative blots in Fig. 4 B show that BAY 11-7082 induced a considerable decrease in the activity of NF-κB in K562-Vinc cells for all tested concentrations. However, the activity of NF-κB was only affected at 20 μM BAY 11-7082 in wild type K562 cells (n = 3). Moreover, we obtained comparable results when assessing the expression of p65 in nuclear extracts by Western blot (data not shown).

In order to confirm that the reduced proliferation observed above was associated to the induction of apoptosis, cells were exposed to imatinib in the absence or presence of 2.5 μM BAY 11-7082 for 24 h and cell death was determined by annexin V. K562-WT and K562-Vinc were incubated with imatinib at equipotent concentration (corresponding to their IC50 and IC75). We observed that BAY 11-7082 sensitized multidrug-resistant K562 cells to cell death and the co-incubation with imatinib increased the percentage of annexin V positive cells ( Fig. 5 A). As an example, the percentage of annexin V positive K562-Vinc cells at a concentration of imatinib close to its IC50, increased ∼70 percent if compared with the treatment with BAY 11-7082 alone. However, no differences were observed in the percentage of apoptosis for wild type K562 cells when these cells were exposed to BAY 11-7082 and/or imatinib during 24 h (p > 0.05, n = 3). In addition, Fig. 5 B shown a representative Western blot analysis of total lysates obtained from K562-WT and K562-Vinc cells exposed to imatinib and/or 2.5 μM BAY 11-7082 during 24 h. These results revealed that cell death induced by both drugs in K562-Vinc cells was preceded by activation of procaspase 9 and procaspase 3, while no changes were observed in K562-WT (n = 3).


Fig. 5 (A) K562-WT and K562-Vinc were exposed to equipotent concentration of imatinib with or without 2.5 μM BAY for 24 h; after which, the extent of apoptosis was determined by annexin V-FITC staining as described in Section 2 . (B) Representative Western blot analysis in K56-WT and K562-Vinc lysates after the treatment with equipotent concentrations of imatinib (0.5 and 5.0 μM for K562-WT and K562-Vinc, respectively) and/or 2.5 μM BAY 11-7082. The antibodies recognize the uncleaved procaspase 3 and procaspase 9. This experiment is representative of three.

3.5. Effect of BAY 11-7082 over P-gp expression

The results presented above suggest that the down-regulation of NF-κB could be involved in the P-gp expression in K562-Vinc cells which over-express this protein. In order to explore this possibility, the expression of P-gp after BAY 11-7082 incubation at the indicated times was evaluated in K562-Vinc protein extracts; and a representative immunoblot is shown in Fig. 6 . We observed a significant down-regulation of P-gp after 72 h of incubation with the NF-κB inhibitor (p < 0.05, n = 3). The percentage of reduction of P-gp expression was 75.6 ± 6.7%.


Fig. 6 (A) Representative Western blot analysis in K562-Vinc lysates. The expression of P-gp after 2.5 μM BAY 11-7082 incubation was analyzed at the indicated times and α-tubulin was used as internal control. (B) The bar graph shows the relative level of P-gp (mean ± S.D.) from three separate experiments. The asterisk indicates statistical significance regarding control.

4. Discussion

The development of imatinib has revolutionized the treatment of patients with CML, although eventually relapse was observed in clinical trials. A study after 5 years of follow-up showed an estimated relapse rate of 17% and 7% of all patients progressed to the accelerated phase [30] . In this sense, several mechanisms of resistance were described, including amplification or mutation of the bcr-abl gene and drug efflux [21], [22], [23], and [31].

We found that the vincristine-selected K562 cell line that exhibited a functional P-gp over-expression showed a MDR phenotype by developing cross-resistance to structurally unrelated drugs, including imatinib. The resistance to imatinib was significantly reversed in the presence of the P-gp inhibitor verapamil. Despite the fact that we cannot rule out a very low expression of P-gp in K562-WT, we have not observed a significant effect of verapamil on the imatinib cytotoxicity in these cells. Our finding is consistent with previous reports that demonstrated that P-gp over-expression confers resistance to imatinib in different types of cells [23], [29], [31], [32], [33], and [34] but not with the observations of Ferrao et al. [35] . Moreover, Galimberti et al. [36] reported that MDR1 expression would play an important role in imatinib resistance, especially when the disease is not fully controlled. Several studies also suggest the involvement of other transporters in the mechanism of imatinib-resistance like BCRP (ABCG2) and hOTC1 [34] and [37]. Although we did not explore the level of these proteins in our model, it has been recently reported that P-gp, but not BCRP and hOCT1, is involved in imatinib-resistance in K562 cells [38] .

It was established that several malignant haematological cells presented a constitutive activation of NF-κB transcription factor [15], [16], and [28]. In accordance with this, we found that NF-κB was constitutively activated in both cell lines although wild type K562 cells exhibited a higher activity compared to drug-resistant cells. It is known that antineoplastic agents promote the activation of NF-κB when they are used in acute conditions [43] even though the information originated from the use of those drugs in the long term is scarce. Although we found no difference by removing the vincristine from de culture medium 24 h before the experiments, we cannot discard an irreversible effect in K562-Vinc cells caused by the permanent exposure to this drug that somehow could down-regulates the NF-κB pathway. It has been described that cellular resistance to vincristine correlated to suppression of NF-κB activation in U937 cell line [39] . Moreover, Yin et al. [40] observed that daunorubicin-resistant K562 cells, which showed minor levels of nuclear p65 than wild type cells, were more susceptible to inhibition of proliferation and induction of apoptosis by forskolin. In contrast, it was reported a higher constitutive NF-κB DNA-binding in chemoresistant cells derived from human ovarian adenocarcinoma cells and murine leukemia T-cells [41] and [42].

It is well known that most of the NF-κB target genes have anti-apoptotic functions and different evidences indicate that inhibition of NF-κB activation may provide a molecular approach to increase apoptosis sensitivity in anticancer treatment [44], [45], and [46]. In this study, we observed that NF-κB inhibition by BAY 11-7082 interfered with basal level of NF-κB activation in multidrug-resistant K562 cells, sensitizing these cells to cell death. Further, BAY 11-7082 increased the imatinib induced cytotoxicity in resistant variant and this effect was associated with activation of procaspases 9 and 3 in K562-Vinc cells indicating a mitochondrial pathway and caspase-dependent mechanism. Although the wild type K562 cells have a constitutive high NF-κB activity, they are sensitive to the IκB kinase (IKK) inhibitor BAY 11-7082, and as expected, higher inhibitor concentrations are required in order to decrease the NF-κB activity. High constitutive NF-κB activity in the wild type cells could be related to the presence of bcr-abl; however, something additional that comprises a difference between both cell variants should be considered. In this regard, there are several transduction signals besides the single activation of IKK that are involved in the control of NF-κB activity. These signals were not investigated in this particular work but could explain the differences between the two cell variants concerning the levels of NF-κB activity. It has been previously demonstrated that cAMP could regulate the activity of NF-κB and the differences in apoptosis sensitivity between both the parental and the resistant K562 cells have been associated to dissimilar levels of cAMP [40] . In addition, NF-κB activity is regulated by the recruitment of coactivators, being RAC3 one of them [47] . Although this molecule is usually over-expressed in breast and ovary tumors, it has been recently described that it is also over-expressed in wild type K562 cells [48] . The levels of RAC3 in the resistant cells remains to be determined, however, this coactivator or additional members of his family may constitute a difference between both cell variants that could explain the dissimilar response to BAY 11-7082. Moreover, our results also suggest that a constitutive NF-κB activity in resistant cells is not strong enough in order to protect cells from apoptosis and could be requiring additional signals such as is observed in wild type cells.

According with previous reports which point out that NF-κB may play different roles in the regulation of human MDR1 gene [7], [8], and [49], our data also indicate that NF-κB could be functionally associated to P-gp since BAY 11-7082 down-regulates the expression of this protein. The down-regulation of P-gp was observed after 72 h of incubation with BAY 11-7082 and this result is consistent with the large turnover previously reported for this protein [50] and [51]. Therefore, we speculate that BAY 11-7082 may reduce the intracellular levels of imatinib since it has been reported that P-gp over-expressing cells have lower intracellular concentration of this drug [32] and [33]. Although our results indicate that verapamil and BAY 11-7082 may be working in the same pathway, we observed an additive effect of these drugs over imatinib cytotoxicity. This is not an unexpected result because these drugs act at different levels of this pathway and the NF-κB inhibitor failed to completely abolish the expression of P-gp. Interestingly, García et al. [42] found that leukemic murine-resistant cells were more sensitive to the activity of BAY 11-7082 via increased apoptosis, although these cells expressed higher constitutive NF-κB activity than the wild type counterpart. On the other hand, Cilloni et al. [52] demonstrated that the inhibition of NF-κB with PS1145 was able to reduce the proliferation of CML cell lines sensitive and resistant to imatinib by continuous exposition to this drug in the culture medium. Given that we observed a similar effect only in MDR cells, it is possible that this discrepancy could be due to the different inhibitor used and different resistant model studied. However, they also observed that imatinib enhances the effect of PS1145 with a further increase of apoptosis and inhibition of proliferation in resistant K562 cells rather than the parental cells. In view of the fact that K562-WT cells resulted less sensitive than resistant cells to BAY 11-7082 effect in our experimental conditions, therefore, we could not speculate concerning the role of NF-κB activity over imatinib action in wild type cells.

In conclusion, we present evidence that the over-expression of functional P-gp is accompanied with a development of MDR in human chronic myeloid leukemia K562 cell line. This MDR phenotype conferred imatinib resistance which was reversed by inhibition of NF-κB pathway. These results suggest that NF-κB could be playing a key role as an anti-apoptotic factor in cells with MDR phenotype and this effect would be independent of the resistance mechanism developed. Moreover, this knowledge could be particularly useful for the development in the future of more effective therapies for CML with MDR phenotype.

Conflict of interest statement



We thank Dr. Lehne G. (University of Oslo, Norway) for the gift of vincristine-resistant K562 cells and Dr. Larripa I. (Academia Nacional de Medicina, Argentina) for the bcr-abl determination. This work was supported in part by grants from National Council of Research of Argentina (CONICET) and University of Buenos Aires, Argentina. F.R., G.C. and S.d.M. are fellows of the CONICET. Y.A.M.C. and B.A.K. are established investigators of CONICET.

Contributions: Yanina Assef contributed to the concept and design, interpreted and analyzed the data, supplied statistical expertise, collected and assembled the data. F. Rubio, G. Coló and S. del Mónaco provided study material, collected and assembled the data. M. Costas and B.A. Kotsias contributed to the concept and design, interpreted and analyzed the data, provided drafting of the article, obtained a funding source, collected and assembled the data.


  • [1] W.D. Stein. Kinetics of the multidrug transporter (P-glycoprotein) and its reversal. Physiol Rev. 1997;77:545-590
  • [2] M.I. Borges-Walmsley, K.S. McKeegan, A.R. Walmsley. Structure and function of efflux pumps that confer resistance to drugs. Biochem J. 2003;376:313-338 Crossref.
  • [3] T.W. Loo, D.M. Clarke. Recent progress in understanding the mechanism of P-glycoprotein-mediated drug efflux. J Membr Biol. 2005;206:173-185 Crossref.
  • [4] Z. Trnková, R. Bedrlíková, J. Marková, K. Michalová, P. Stöckbauer, J. Schwarz. Semiquantitative RT-PCR evaluation of the MDR1 gene expression in patients with acute myeloid leukaemia. Neoplasma. 2007;54:383-390
  • [5] J.C. Cusack Jr., R. Liu, A.S. Baldwin Jr. Inducible chemoresistance to 7-ethyl-10-[4-(1-piperidino)-1-piperidino]-carbonyloxycamptothecin (CPT-11) in colorectal cancer cells and a xenograft model is overcome by inhibition of nuclear factor-kappaB activation. Cancer Res. 2000;60:2323-2330
  • [6] J.H. Um, C.D. Kang, B.G. Lee, D.W. Kim, B.S. Chung, S.H. Kim. Increased and correlated nuclear factor-kappa B and Ku autoantigen activities are associated with development of multidrug resistance. Oncogene. 2001;20:6048-6056 Crossref.
  • [7] M.T. Kuo, Z. Liu, Y. Wei, et al. Induction of human MDR1 gene expression by 2-acetylaminofluorene is mediated by effectors of the phosphoinositide 3-kinase pathway that activate NF-kB signalling. Oncogene. 2002;21:1945-1954 Crossref.
  • [8] M. Bentires-Alj, V. Barbu, M. Fillet, A. Chariot, B. Relic, N. Jacobs, et al. NF-kappaB transcription factor induces drug resistance through MDR1 expression in cancer cells. Oncogene. 2003;22:90-97 Crossref.
  • [9] A.S. Baldwin. Series introduction: the transcription factor NF-kappaB and human disease. J Clin Invest. 2001;107:3-6 Crossref.
  • [10] J.W. Pierce, R. Schoenleber, G. Jesmok, et al. Novel inhibitors of cytokine-induced IkappaBalpha phosphorylation and endothelial cell adhesion molecule expression show anti-inflammatory effects in vivo. J Biol Chem. 1997;272:21096-21103 Crossref.
  • [11] Y. Dai, M. Rahmani, P. Dent, S. Grant. Blockade of histone deacetylase inhibitor-induced RelA/p65 acetylation and NF-kappaB activation potentiates apoptosis in leukemia cells through a process mediated by oxidative damage, XIAP downregulation, and c-Jun N-terminal kinase 1 activation. Mol Cell Biol. 2005;25:5429-5444 Crossref.
  • [12] P. Bauerle, D. Baltimore. NF-κB: ten years after. Cell. 1996;87:13-20
  • [13] S. Ghosh, M. Karin. Missing pieces in the NF-kappaB puzzle. Cell. 2002;109:S81-S96 Crossref.
  • [14] M. Karin. Nuclear factor-kB in cancer development and progression. Nature. 2006;44:431-436 Crossref.
  • [15] U. Kordes, D. Krappmann, V. Heissmeyer, W.D. Ludwig, C. Scheidereit. Transcription factor NF-kappaB is constitutively activated in acute lymphoblastic leukemia cells. Leukemia. 2000;14:399-402 Crossref.
  • [16] M. Guzman, S. Neering, D. Upchurch, et al. Nuclear factor-κB is constitutively activated in primitive human acute myelogenous leukemia cells. Blood. 2001;98:2301-2307 Crossref.
  • [17] T. Sanda, S. Iida, H. Ogura, et al. Growth inhibition of multiple myeloma cells by a novel IKB kinase inhibitor. Clin Cancer Res. 2005;11:1974-1982 Crossref.
  • [18] C.B. Lozzio, B.B. Lozzio. Human chronic myelogenous leukemia cell line with positive Philadelphia chromosome. Blood. 1977;45:321-334
  • [19] B.J. Druker, S. Tamura, E. Buchdunger, et al. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med. 1996;2:561-566 Crossref.
  • [20] M. Deininger, E. Buchdunger, B.J. Druker. The development of imatinib as a therapeutic agent for chronic myeloid leukemia. Blood. 2005;105:2640-2653 Crossref.
  • [21] F.X. Mahon, M.W. Deininger, B. Schultheis, et al. Selection and characterization of BCR-ABL positive cell lines with differential sensitivity to the tyrosine kinase inhibitor STI571: diverse mechanisms of resistance. Blood. 2000;96:1070-1079
  • [22] P. le Coutre, E. Tassi, M. Varella-Garcia, et al. Induction of resistance to the Abelson inhibitor STI571 in human leukemic cells through gene amplification. Blood. 2000;95:1758-1766
  • [23] F.X. Mahon, F. Belloc, V. Lagarde, et al. MDR1 gene overexpression confers resistance to imatinib mesylate in leukemia cell line models. Blood. 2003;101:2368-2373 Crossref.
  • [24] Y. Assef, S. Cavarra, A. Damiano, C. Ibarra, B.A. Kotsias. Ionic currents in multidrug resistant K562 human leukemic cells. Leukemia Res. 2005;29:1039-1047 Crossref.
  • [25] M.F. Rubio, S. Werbajh, E.G. Cafferata, et al. TNF-alpha enhances estrogen-induced cell proliferation of estrogen-dependent breast tumor cells through a complex containing nuclear factor-kappaB. Oncogene. 2006;25:1367-1377 Crossref.
  • [26] G. Lehne, P. De Angelis, M. den Boer, H.E. Rugstad. Growth inhibition, cytokinesis failure and apoptosis of multidrug-resistant leukemia cells after treatment with P-glycoprotein inhibitory agents. Leukemia. 1999;13:768-778 Crossref.
  • [27] T. Fukushima, T. Yamashita, H. Takemura, H. Suto, S. Kishi, Y. Urasaki, et al. Effect of PSC 833 on the cytotoxicity and pharmacodynamics of mitoxantrone in multidrug-resistant K562 cells. Leukemia Res. 2000;24:249-254 Crossref.
  • [28] B.M. Pickering, S. de Mel, M. Lee, et al. Pharmacological inhibitors of NF-kappaB accelerate apoptosis in chronic lymphocytic leukaemia cells. Oncogene. 2007;26:1166-1177 Crossref.
  • [29] M. Mukai, X.F. Che, T. Furukawa, et al. Reversal of the resistance to STI571 in human chronic myelogenous leukemia K562 cells. Cancer Sci. 2003;94:557-563 Crossref.
  • [30] B.J. Druker, F. Guilhot, S.G. O’Brien, et al. Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med. 2006;355:2408-2417 Crossref.
  • [31] J. Thomas, L. Wang, R.E. Clark, M. Pirmohamed. Active transport of imatinib into and out of cells: implications for drug resistance. Blood. 2004;104:3739-3745 Crossref.
  • [32] T. Illmer, M. Schaich, U. Platzbecker, J. Freiberg-Richter, U. Oelschlägel, M. von Bonin, et al. P-glycoprotein-mediated drug efflux is a resistance mechanism of chronic myelogenous leukemia cells to treatment with imatinib mesylate. Leukemia. 2004;18:401-408 Crossref.
  • [33] N. Widmer, H. Rumpold, G. Untergasser, A. Fayet, T. Buclin, L.A. Decosterd. Resistance reversal by RNAi silencing of MDR1 in CML cells associated with increase in imatinib intracellular levels. Leukemia. 2007;21:1561-1562 Crossref.
  • [34] H. Rumpold, A.M. Wolf, K. Gruenewald, G. Gastl, E. Gunsilius, D. Wolf. RNAi-mediated knockdown of P-glycoprotein using a transposon-based vector system durably restores imatinib sensitivity in imatinib-resistant CML cell lines. Exp Hematol. 2005;33:767-775 Crossref.
  • [35] P.T. Ferrao, M.J. Frost, S.P. Siah, L.K. Ashman. Overexpression of P-glycoprotein in K562 cells does not confer resistance to the growth inhibitory effects of imatinib (STI571) in vitro. Blood. 2003;102:4499-4503 Crossref.
  • [36] S. Galimberti, G. Cervetti, F. Guerrini, et al. Quantitative molecular monitoring of BCR-ABL and MDR1 transcripts in patients with chronic myeloid leukemia during Imatinib treatment. Cancer Genet Cytogenet. 2005;162:57-62 Crossref.
  • [37] H. Burger, H. van Tol, M. Brok, E.A. Wiemer, E.A. de Bruijn, G. Guetens, et al. Chronic imatinib mesylate exposure leads to reduced intracellular drug accumulation by induction of the ABCG2 (BCRP) and ABCB1 (MDR1) drug transport pumps. Cancer Biol Ther. 2005;4:747-752 Crossref.
  • [38] C. Hirayama, H. Watanabe, R. Nakashima, T. Nanbu, A. Hamada, A. Kuniyasu, et al. Constitutive overexpression of P-glycoprotein, rather than breast cancer resistance protein or organic cation transporter 1, contributes to acquisition of imatinib-resistance in K562 cells. Pharm Res. 2008;25:827-835 Crossref.
  • [39] D.K. Giri, P. Pantazis, B.B. Aggarwal. Cellular resistance to vincristine suppresses NF-kappa B activation and apoptosis but enhances c-Jun-NH2-terminal protein kinase activation by tumor necrosis. Apoptosis. 1999;4:291-301 Crossref.
  • [40] Y. Yin, P.D. Allen, L. Jia, M.G. MacEy, S.M. Kelsey, A.C. Newland. Constitutive levels of cAMP-dependent protein kinase activity determine sensitivity of human multidrug-resistant leukaemic cell lines to growth inhibition and apoptosis by forskolin and tumour necrosis factor alpha. Br J Haematol. 2000;108:565-573 Crossref.
  • [41] C. Salvatore, G. Camarda, C.A. Maggi, C. Goso, S. Manzini, M. Binaschi. NF-kappaB activation contributes to anthracycline resistance pathway in human ovarian carcinoma cell line A2780. Int J Oncol. 2005;27:799-806
  • [42] M.G. García, L. Alaniz, E.C. Lopes, G. Blanco, S.E. Hajos, E. Alvarez. Inhibition of NF-kappaB activity by BAY 11-7082 increases apoptosis in multidrug resistant leukemic T-cell lines. Leukemia Res. 2005;29:1425-1434
  • [43] K.C. Das, C.W. White. Activation of NF-kappaB by antineoplastic agents. Role of protein kinase C. J Biol Chem. 1997;272:14914-14920 Crossref.
  • [44] A. Beg, D. Baltimore. An essential role for NF-κB in preventing TNF-α-induced cell death. Science. 1996;274:782-784 Crossref.
  • [45] I. Jeremias, C. Kupatt, B. Baumann, I. Herr, T. Wirth, K.M. Debatin. Inhibition of nuclear factor κB activation attenuates apoptosis resistance in lymphoid cells. Blood. 1998;91:4624-4631
  • [46] H.J. Kim, N. Hawke, A.S. Baldwin. NF-kappaB and IKK as therapeutic targets in cancer. Cell Death Differ. 2006;13:738-747 Crossref.
  • [47] S. Werbajh, I. Nojek, R. Lanz, M.A. Costas. RAC-3 is a NF-κB coactivator. FEBS Lett. 2000;485:195-199 Crossref.
  • [48] G. Colo, R. Rosato, S. Grant, M.A. Costas. RAC3 down-regulation sensitizes human chronic myeloid leukemia cells to TRAIL-induced apoptosis. FEBS Lett. 2007;581:5075-5081 Crossref.
  • [49] B. Ogretmen, A. Safa. Negative regulation of MDR1 promoter activity in MCF-7, but not in multidrug resistant MCF-7/Adr, cells by cross-coupled NF-kB/p65 and c-Fos transcription factors and their interaction with the CAAT region. Biochemistry. 1999;38:2189-2199 Crossref.
  • [50] W. Zhang, V. Ling. Cell-cycle-dependent turnover of P-glycoprotein in multidrug-resistant cells. J Cell Physiol. 2000;184:17-26 Crossref.
  • [51] C. Muller, G. Laurent, V. Ling. P-glycoprotein stability is affected by serum deprivation and high cell density in multidrug-resistant cells. J Cell Physiol. 1995;163:538-544 Crossref.
  • [52] D. Cilloni, F. Messa, F. Arruga, et al. The NF-kappaB pathway blockade by the IKK inhibitor PS1145 can overcome imatinib resistance. Leukemia. 2006;20:61-67 Crossref.


a Laboratorio de Neurofisiología, Instituto de Investigaciones Médicas Alfredo Lanari, Universidad de Buenos Aires, Conicet, Argentina

b Laboratorio de Biología Molecular y Apoptosis, Instituto de Investigaciones Médicas Alfredo Lanari, Universidad de Buenos Aires, Conicet, Argentina

lowast Corresponding author at: Inst. Inv. Médicas A. Lanari, C. de Malvinas 3150, 1427 Buenos Aires, Argentina.