Introduction

Chronic myelogenous leukemia accounts for 15% of adult leukemias. In 2009, an estimated 5,050 cases will be diagnosed in the USA, and 470 patients will die from the disease [1]. Although adriamycin has high efficacy against chronic myelogenous leukemia, resistance to adriamycin is a major clinical problem and an important cause for treatment failure. Several mechanisms have been suggested to cause resistance to adriamycin in chronic myelogenous leukemia. They include the upregulation of P-glycoprotein and other plasma membrane multidrug transporters [2, 3]; failure to undergo apoptosis caused by alterations in Bcl-2, Bcl-XL, or Bax expression [4–6]; and changes in drug targets, such as topoisomerase II [7]. But treatments targeting these pathways have been met with little success. Thus, identification of novel diagnostic and therapeutic markers associated with adriamycin-resistance of tumor is very meaningful.

New methods are emerging that facilitate identification of novel diagnostic and therapeutic markers for cancer. In particular, the proteomic approach has resulted in many opportunities and challenges in identifying new tumor markers and therapeutic targets, and in understanding disease pathogenesis [8]. The most consistently successful proteomic method is combination of two-dimensional gel electrophoresis (2-DE) for protein separation and visualization, then mass spectrometric (MS) identification of proteins using peptide mass fingerprints and tandem MS peptide sequencing [9]. Although it has some limitations (for example, labor intensive and the limited dynamic range of proteins), this approach has been used extensively in cancer research [9]. Two dimensional gel electrophoresis-based comparative proteome analysis has been seen as useful for identification of resistance-associated proteins and has been successfully used for screening potential biomarkers for drug resistance in tumor cell lines, serum, and tumor tissues [10–13].

Recently, Zhu et al. compared the global protein profiles between adriamycin-resistant K562 cells and the parental K562 cells, and found eight differentially expressed proteins [14]. However, chemo-resistance acquisition is a very complicated process, and there must be many additional factors that play important roles in the complex processes. And the aim of the present study was to find more proteins involved in adriamycin-resistance using proteomic tools so as to develop more strategies to overcome drug resistance. To accomplish this, the cell line K562 was treated continuously, over 6 months, with stepwise-increasing concentrations of adriamycin to yield the chemo-resistant cell line that was named K562/ADM. The resistance of K562/ADM cells to adriamycin was 239-fold higher than the parental K562 cells, which showed higher resistance than the cells established by Zhu et al. (149-fold increased resistance to ADM) [14]. When subject the two cell lines to proteomic investigation, 29 significantly changed proteins were identified by 2-DE and MS/MS analysis. Altered expression of Annexin A2, Annexin A4, and Phosphoglycerate kinase 1 proteins were further verified by Western blot. The data in the present study may facilitate a better understanding of the molecular mechanisms underlying adriamycin resistance of K562 cells.

Materials and methods

Cell culture

The human chronic myelogenous leukemia cell K562 cells were maintained in RPMI 1640 medium (Gibco, Grand Island, NY) with 100 U/ml penicillin, 100 g/ml streptomycin, and supplemented with 10% fetal calf serum (Hyclone, Logan, UT) at 37°C in a humidified atmosphere of 5% CO2. A multidrug resistant cell line K562/ADM was established by continuous exposure to stepwise increased dosage of adriamycin (Hisun Pharmaceuticals, China). Briefly cells were exposed to adriamycin starting from 1 × IC50, and the concentrations were escalated at an increasing rate of 50%. Finally, the cells were maintained in the above medium with 1 g/ml adriamycin. For 2-DE analysis, cells (80% confluence) were washed twice with PBS before harvesting. The cells were stored at −80°C until use.

Drug sensitivity assay

Cultured cells were suspended in RPMI 1640 at 6 × 104 cells/ml and 100 μl of cell suspension were seeded into a well of 96-well plate. After an overnight culture, cells were exposed to adriamycin for 48 h. After incubation with MTT (1 mg/ml) for 4 h, medium were removed and 200 μl of dimethylsulfoxide were added to dissolve formazan crystals. Absorbance was measured at 570 nm using an ELISA microplate reader (PerkinElmer, Wellesley, MA, USA). Cell survival rates were calculated according to the formula: survival rate = (mean O.D. of treated wells/mean O.D. of untreated wells) × 100%. IC50 values were determined by drawing dose–effect curves on semi logarithm coordinate paper.

2-DE and image analysis

The 2-DE was performed as described previously [15] with minor modification. Approximately 4 × 107 cells were lysed in 1 ml lysis buffer (7 M urea, 2 M thiourea, 4% CHAPS, 100 mM DTT, 0.2% pH 3–10 ampholyte, Bio-Rad, USA) containing protease inhibitor cocktail 8340 (Sigma, St. Louis, MO). Samples were then kept on ice and sonicated in six cycles of 15 s, each consisting of 5 s sonication followed by a 10 s break. After centrifugation at 15,000 rpm for 1 h at 4°C, the supernatant was collected and the protein concentrations were determined using the DC protein assay kit (Bio-Rad). Protein samples (2 mg) were applied to IPG strip (17 cm, pH 3–10 NL, Bio-Rad) using a passive rehydration method. After 12–16 h of rehydration, the strips were transferred to an IEF Cell (Bio- Rad). IEF was performed as follows: 250 V for 30 min, linear; 1,000 V for 1 h, rapid; linear ramping to 10 000 V for 5 h and finally 10,000 V for 4 h. The second dimension was performed using 12% SDS-PAGE at 30 mA constant current per gel after equilibration. The gels were stained using CBB R-250 (Merck, Germany) and scanned with a Bio-Rad GS-800 scanner. Three independent runs were made for each cell lines to ensure the accuracy of analyses. The maps were analyzed by PDQuest software Version 6.1 (Bio-Rad). The quantity of each spot in a gel was normalized as a percentage of the total quantity of all spots in that gel and evaluated in terms of OD. For statistical analysis, paired t test was performed to compare data from the three repeated experiments. Only spots that showed consistent and significant differences (±over twofold, P < 0.05) were selected for analysis with MS.

In-gel digestion

In-gel digestion of proteins was carried out using MS-grade Trypsin Gold (Promega, Madison, WI) according to the manufacturer’s instructions. Briefly, spots were cut out of the gel (1–2 mm diameter) using a razor blade, and destained twice with 100 mM NH4HCO3/50% ACN at 37°C for 45 min in each treatment. After drying, the gels were preincubated in 10–20 μl trypsin solution for 1 h. Then, 15 μl digestion buffer was added (40 mM NH4HCO3/10% ACN) to cover each gel and incubated overnight at 37°C. Tryptic digests were extracted using MilliQ water initially, followed by two times extraction with 50% ACN/5% TFA for 1 h each time. The combined extracts were dried in a vacuum concentrator at room temperature. The samples were then subjected to MS analysis.

ESI-Q-TOF analysis and protein identification

Mass spectra were acquired using a Q-TOF mass spectrometer (Micromass, Manchester, UK) fitted with an ESI source (Waters). Tryptic digests were dissolved in 18 μl 50% ACN. MS/MS was performed in a data-dependent mode in which the top 10 most abundant ions for each MS scan were selected for MS/MS analysis. Trypsin autolysis products and keratin derived precursor ions were automatically excluded. The MS/MS data were acquired and processed using MassLynx software (Micromass) and MASCOT (http://www.matrixscience.com) was used to search the database. Database searches were carried out using the following parameters: Database, Swiss-Prot; taxonomy, homo sapiens; enzyme, trypsin; mass tolerance, ±0.1 Da; MS/MS tolerance, ±0.05 Da; and an allowance of one missed cleavage. Fixed modifications of cysteine carboamidomethylation, and variable modifications of methionine oxidation were allowed. The data format was selected as Micromass PKL and the instrument was selected as ESI-Q-TOF. Proteins with probability based MOWSE scores, derived from ions scores as a non-probabilistic basis for ranking protein hits when using Mascot searching engine, exceeding their threshold (P < 0.05) were considered to be positively identified. To eliminate the redundancy of proteins appearing in the database under different names or accession numbers, the one-protein member with the highest MASCOT score was further selected from the relevant multiple-member protein family.

Western blot

K562 and K562/ADM cell proteins were extracted in RIPA buffer (50 mM Tris-base, 1.0 mM EDTA, 150 mM NaCl, 0.1% SDS, 1% Triton X-100, 1% Sodium deoxycholate, 1 mM PMSF) and quantified by the DC protein assay kit (Bio-Rad). Samples were separated by 12% SDS-PAGE and transferred to PVDF membranes (Amersham Biosciences). The membranes were blocked overnight with PBS containing 0.1% Tween 20 in 5% skimmed milk at 4°C, and subsequently probed by the primary antibodies: rabbit-anti-Annexin A2 (diluted 1:500 Santa Cruz Biotechnology, USA), rabbit-anti-Annexin A4 (diluted 1:1,000 Santa Cruz Biotechnology, USA), and rabbit-anti-Phosphoglycerate kinase 1 (diluted 1:500 Santa Cruz Biotechnology, USA). Blots were incubated with the respective primary antibodies for 2 h at room temperature and washed three times in TBST. After that, the blots were incubated with secondary antibody (diluted 1:5,000) conjugated to HRP for 2 h at room temperature. Target proteins were detected by enhanced chemiluminescence reagents (Amersham Pharmacia Biotech, Piscataway, USA). GAPDH was used as an internal control.

Results

K562/ADM cell line acquirement and 2-DE analysis

Adriamycin-resistant cell line K562/ADM was established by treating continuously, over 6 months, with stepwise-increasing concentration of adriamycin. The IC50 values for adriamycin treatment in K562 and K562/ADM cells were 0.06 ± 0.04 and 14.36 ± 1.08 μM, respectively. The resistance of K562/ADM cells to adriamycin was 239-fold higher than that of the parental K562 cells, which meant adriamycin-resistant cell line was successfully established. The protein expression profile of K562/ADM and K562 cells were examined by 2-DE. The experiments were repeated three times. A pair of representative 2-DE maps was shown in Fig. 1. The spots with remarkable expression changes were selected as examples and shown in Fig. 2 in an enlarged form.

Fig. 1
figure 1

Comparison of the protein expression patterns between K562 cells and adriamycin-resistant K562 cells. The arrows indicate the 29 differentially expressed proteins

Full size image
Fig. 2
figure 2

Enlargement of selected regions as examples showing spots with consistent expression variation in 2-DE gels. a Spots # 1, # 2, and # 4 (corresponding to ANXA4, PSMD8, and PDAP1, respectively) were significantly up-regulated in K562/ADM cells compared with K562 cells. b The quantity of each spot was determined by PDQuest software. # 2 spot was up-regulated 3.6-fold ± 1.3 (P < 0.05) in K562/ADM cells compared with control K562 cells; # 1 and # 4 were also significantly up-regulated (P < 0.01). Data were shown as mean values ± SD

Full size image

Identification of differentially expressed proteins

Out of the 35 differentially expressed proteins in pair wise comparison between the two cell lines, 29 proteins were successfully identified using ESI-Q-TOF MS/MS analysis (Table 1). These proteins can be grouped into 10 main functional classes including metabolism (24%), proteolysis (13%), signal transduction (21%) and calcium ion binding (6%), and so on (Fig. 3). Representative MS maps of spot #8 were shown in Fig. 4. The MOWSE score and the number of unique peptides identified by MS/MS sequencing were 685 and 33, respectively (Table 1).

Table 1 Identified proteins by MS/MS analysis
Full size table
Fig. 3
figure 3

29 differentially expressed proteins between K562/ADM cells and K562 cells were functionally classified into 10 groups, including metabolism (24%), proteolysis (13%), signal transduction (21%), and calcium ion binding (6%)

Full size image
Fig. 4
figure 4

Results of ESI-Q-TOF MS/MS identification of representative protein Annexin A2. a Mass spectrogram of tryptic peptides from spot #8. b An example of an MS/MS spectrum of parent ion 826.0068. c Protein sequence of Annexin A2. The matched peptides are underlined

Full size image

Protein validation by Western blot

The alteration of Annexin A2, Annexin A4 and Phosphoglycerate kinase 1 proteins were further validated by Western blot. Consistent with the observations in 2-DE analysis, Annexin A2, Annexin A4 and Phosphoglycerate kinase 1 were found up-regulated in the K562/ADM compared with the parental K562 cells (Fig. 5).

Fig. 5
figure 5

Western blot confirmation of the alterations of Annexin A2, Annexin A4, and Phosphoglycerate kinase 1 proteins. The images were representative results of at least three independent experiments. Annexin A2, Annexin A4, and Phosphoglycerate kinase 1 proteins were significantly up-regulated (P < 0.01) in K562/ADM cells compared with K562 cells. GADPH was used as a loading control

Full size image

Discussion

Acquired resistance to multiple anticancer drugs is a major problem in chemotherapy, and some mechanisms of resistance have been investigated previously [2–7]. However, the exact mechanisms that lead to chemo-resistance are largely unknown. In the present study, a comparative proteome study using 2-DE and MS based analysis was conducted and totally 29 proteins were positively identified to be differentially expressed between K562/ADM and the parental K562 cells.

Compared with the previous study [14], we identified more differentially expressed proteins potentially involved in adriamycin-resistance. Most of identified proteins were related to the categories of metabolism, proteolysis, signal transduction and calcium ion binding, suggesting that alterations of those biological processes might be involved in adriamycin resistance of K562 cells. Among them, some of the proteins have been reported to be associated with the development of tumor drug resistance, such as Annexin A2, while some might be new drug-resistance-associated proteins, such as PGK1, and there are also some proteins that have not been previously linked to drug resistance. The functional roles of the changed proteins associated with drug-resistance of tumor cells were briefly discussed below.

Hypoxia was often observed in solid tumors [16]. The oxygen-deficient hypoxic tumor cells have been shown to influence prognosis in terms of both a negative effect on response to chemotherapy/radiotherapy and a positive effect on malignant progression [17–19]. The mechanisms by which hypoxia contributes to chemoresistance are largely unknown. Tumor hypoxia may develop in tumors with a high growth rate relative to the capacity of the microvasculature to supply oxygen and nutrients, causing cell cycle disruption or arrest, DNA over-replication, and induction of stress proteins [20]. Indeed, our proteomic analysis demonstrated over-expression of glycolytic enzymes such as phosphoglycerate kinase 1, transaldolase, fructose-bisphosphate aldolase A, and l-lactate dehydrogenase B chain, indicating the presence of hypoxia in the chemoresistant tumor cells. Thus, it may be an effective method to deal with the drug resistance by reversing the hypoxia.

The proteasome, a multicatalytic proteinase complex, is responsible for the majority of intracellular protein degradation [21]. Pharmacologic inhibitors of the proteasome possess in vitro and in vivo anti-tumor activity [22, 23]. PS-341 is a dipeptidyl boronic acid derivative that specifically inhibits the function of the 26S proteasome [24]. The 26S proteasome is a large multisubunit ATP-dependent threonine protease complex found in the cytoplasm and nucleus of all eukaryotic cells; its principle function is to degrade proteins via the ubiquitin pathway [25]. Preclinical and clinical studies demonstrated that the PS-341 potentiates the sensitivity of conventional chemotherapeutic agents, in part by down-regulating chemo-resistance pathways [26]. In this study, we found that some subunits of proteasome including 26S proteasome non-ATPase regulatory subunit 8, 26S proteasome non-ATPase regulatory subunit 11, 26S proteasome non-ATPase regulatory subunit 13 and proteasome subunit beta type-4 were obviously up-regulated in K562/ADM cells compared with K562 cells. These data showed inhibition of proteasome activity may be an effective strategy to reverse chemoresistance.

The Rho GDP dissociation inhibitor (RhoGDI), which is a member of a family of GDIs that include RhoGDI (also known as RhoGDIa or RhoGDI-1), D4-GDI, and RhoGDI-3, is a cellular regulatory protein that works primarily by controlling the cellular distribution and activity of Rho GTPases [27–29]. It is ubiquitously expressed in cells and tissues and binds to most of the Rho GTPases thus far examined, including Rac1, RhoA, and Cdc42 [30]. The mechanism by which RhoGDI can negatively regulate Rho proteins is as follows: (a) by blocking their interaction with guanine nucleotide exchange factors (GEF), thereby inhibiting GTPase activation; (b) by shielding the membrane-anchoring domain of the GTPases, thereby restricting them to a cytosolic (non-active) localization; and (c) by blocking binding to downstream target molecules (e.g., various kinases regulated by small GTPases). RhoGDI is overexpressed in multiple types of human cancers. Previous studies showed that RhoGDI is highly expressed in stable chemoresistant cancer cell lines, such as melanoma and ovarian cells, raising the possibility that RhoGDI might play a role in the development of drug resistance of cancer cells [31–33]. Moreover, the recent data displayed that overexpression of RhoGDI increased resistance of human breast cancer cell MDA-MB-231 and lymphoma cell JLP-119 to the induction of apoptosis by two chemotherapeutic agents: etoposide and doxorubicin. Conversely, silencing of RhoGDI expression by small interfering RNA sensitized these cells to drug-induced apoptosis. The mechanism is RhoGDI has the ability to inhibit caspase-mediated cleavage of Rac1 GTPase, which is required for maximal apoptosis to occur in response to cytotoxic drugs [34]. Our data displayed that both Rho GDP-dissociation inhibitor 1 and Rho GDP-dissociation inhibitor 2 were upregulated in the adriamycin-resistant cells, which means RhoGDI maybe an ideal target for overcoming the adriamycin-resistance.

Ion binding proteins Annexin A2 and A4 correlate with increasing drug resistance. Annexin A2 was identified as highly expressed in adriamycin-resistant human breast carcinoma cells, but not in adriamycin-sensitive human breast carcinoma cells. Small interference RNA-mediated gene suppression demonstrated that Annexin A2 was required for enhanced cell proliferation and invasion of the adriamycin-resistant human breast carcinoma cells. Down-regulation of Annexin A2 alone was not sufficient to reverse the cell sensitivity to adriamycin, suggesting that Anxa2 was not required for MDR phenotype [35]. Previous study has also demonstrated elevated levels of Annexin A4 protein is associated with enhanced chemoresistance to carboplatin in human epithelial ovarian cancer cells in part by enhancing drug efflux [36]. In this study, these two proteins were positively identified with high confidence and further validated by Western blot. Taken together, ion binding proteins Annexin A2 and A4 may play important roles in the development of chemo-resistance.

Molecular chaperones are responsive to a variety of stresses such as cytotoxic agents, and therefore comprise a set of proteins with the potential to be used as biomarkers in response to cytotoxic agents. Each member of molecular chaperones appears to have a distinct set of functions within the cells. Some of these functions are well documented, such as the modulatory effects of HSPs on apoptosis; others need further elucidation, for example the role of special HSPs in thermotolerance [37, 38]. Previous study reported that HSP70A8 and HSP60, CCT3, and HSPC124 were increased in the taxotere-resistant squamous lung carcinoma cell line. Conversely, in the taxol-resistant squamous lung carcinoma cell line, HSP70A8, HSP27, and HSP70 9B were all down regulated [39]. In our study, some molecular chaperones like CCT7, HSPB1, TCP1, and CCT2 were changed significantly (2.6, 3.3, 3.5, and 3.3 folds, respectively) in the adriamycin-resistant cells compared with parental cells, some of which may be involved in the resistance to adriamycin.

Previously Zhu et al. compared the global protein profiles between adriamycin-resistant K562 cells and the parental K562 cells, and found eight differentially expressed proteins [14]. However, it was surprised that the eight proteins found by Zhu et al. did not appeared in the 29 differentially expressed proteins identified in the present study. The following reasons may partly explain why two similar studies lead to different results. Firstly, as we know, chemo-resistance acquisition is a very complicated process, and chemo-resistant cells are very heterogeneous, and maybe different cells gained chemo-resistance through different mechanisms during long time screening strength. Secondly, the type of IPG strip and concentration of SDS-PAGE used in the present study were non-linear and 12%, respectively, which were different from the study performed by Zhu et al. Thirdly, the resistance to adriamycin of K562/ADM cells established in this study was 239-fold higher than their parental K562 cells, and are different from that used by Zhu et al. (149-fold).

Accumulating data suggests that multiple complex mechanisms may be involved, simultaneously or complementarily, in the emergence and development of drug resistance in cancers. In the present study, using proteomic approach, we identified 29 protein spots that have altered expression in the adriamycin-resistant human chronic myelogenous leukemia cell line K562/ADM compared with the parental sensitive cell line K562, and the alteration of three representative proteins were further confirmed by Western blot. We believe the data in the present study may be useful for further study of resistance mechanisms and screening of resistant biomarkers.