Journal of Cancer Research and Clinical Oncology

, Volume 137, Issue 3, pp 521–532

Comparative proteomic analysis of paclitaxel sensitive A549 lung adenocarcinoma cell line and its resistant counterpart A549-Taxol

Authors

  • Qiang-ling Sun
    • Basic Research LaboratoryShanghai Chest Hospital Affiliated to Shanghai Jiaotong University
    • Basic Research LaboratoryShanghai Chest Hospital Affiliated to Shanghai Jiaotong University
  • Xiao-hua Yang
    • Basic Research LaboratoryShanghai Chest Hospital Affiliated to Shanghai Jiaotong University
  • Guo-liang Bao
    • Basic Research LaboratoryShanghai Chest Hospital Affiliated to Shanghai Jiaotong University
  • Jing Lu
    • Institute of HematologyRuijin Hospital of Shanghai Jiaotong University
  • Yin-yin Xie
    • Institute of HematologyRuijin Hospital of Shanghai Jiaotong University
Original Paper

DOI: 10.1007/s00432-010-0913-9

Cite this article as:
Sun, Q., Sha, H., Yang, X. et al. J Cancer Res Clin Oncol (2011) 137: 521. doi:10.1007/s00432-010-0913-9

Abstract

Purpose

Paclitaxel is used as the first-line chemotherapy for Non-Small Cell Lung Cancer (NSCLC), but acquired resistance becomes a critical problem. Several mechanisms have been proposed in paclitaxel resistance, but they are not sufficient to exhaustively explain this resistance emergence. To better investigate molecular resistance mechanisms, a comparative proteomic approach was carried out to identify differentially expressed proteins between human lung adenocarcinoma A549 cell line (paclitaxel sensitive) and A549-Taxol cell line (acquired resistant).

Methods

A paclitaxel-resistant subline (A549-Taxol) derived from the parental-sensitive cell line A549 was established by stepwise selection by paclitaxel. Total proteins in the two cell lines were separated by fluorescent differential gel electrophoresis (DIGE). Image analysis was carried out with the DeCyder 2D 6.5 software. Proteins associated with chemoresistance process were identified by matrix-assisted laser desorption/ionization time-of-flight tandem mass spectrometry (MALDI-TOF–MS/MS). Some key molecules were valuated by Western blot.

Results

Thirty proteins were identified and grouped into eight main functional classes according to the biological processes in which they are likely to participate, i.e. signal transduction, cytoskeleton, redox reaction, energy and metabolism, and so on. Alterations of these processes might be involved in paclitaxel resistance. Most of the proteins showed mitochondrial and cytoplasm location. The up-regulation of CK8, CK18, ALDH1, CAST and ANX I in A549-Taxol cell line was verified by Western blot, in coincidence with the data obtained from proteomic analysis.

Conclusion

For the first time, differentially expressed proteins between paclitaxel-sensitive cell line and paclitaxel-resistant one were explored by comparative proteomic approach in human lung adenocarcinoma. It may be useful for further studying of resistance mechanisms and screening of resistance biomarkers, so as to develop tailored therapeutic strategies.

Keywords

ChemoresistanceNon-Small Cell Lung Cancer (NSCLC)PaclitaxelTwo-dimensional difference gel electrophoresis (2D-DIGE)

Abbreviations

2-DE

Two-dimensional gel electrophoresis

CK8

Cytokeratin-8

CK18

Cytokeratin-18

G6PD

Glucose-6-phosphate 1-dehydrogenase

RALDH1

Retinal dehydrogenase 1

Introduction

Lung cancer ranks first in cancer mortality among people for more than 1.3 million deaths worldwide annually. Most cases are advanced at diagnosis because lung cancer typically does not cause symptoms until having spread outside the lung (Jemal et al. 2005). Platinum compounds, given as a combination of paclitaxel, are the most active lung adenocarcinoma chemotherapy and standard treatment for nearly all patients diagnosed with NSCLC (Rowinsky and Donehower 1995). Although most patients initially respond well to this treatment, the efficiency of chemotherapy is weakened due to the appearance of paclitaxel resistance (Yusuf et al. 2003).

Proteins differentially expressed in paclitaxel-resistant and -sensitive lung adenocarcinoma are likely to be logical candidates for treatment response biomarkers and therapeutic targets (Seve and Dumontet 2005). Proteomics is a potent strategy to provide insights into global protein changes of paclitaxel resistance, which considered to be multifactorial. Two-dimensional difference gel electrophoresis (2D-DIGE) is an important tool in proteomics, allowing the resolution of thousands of protein spots, resulting in a global view of the proteome (Timms and Cramer 2008). DIGE involves labeling samples with spectrally resolvable fluorescent CyDyes prior to electrophoresis. All samples are mixed before isoelectric focusing and resolved on the same 2-DE gel. This methodology dramatically reduces variation in spot intensities due to gel-specific experimental factors, as the effects will be the same for each sample within a single DIGE gel (Lilley 2003).

In the present study, to gain further insight into the mechanism underlying paclitaxel resistance, we established a paclitaxel-resistant human lung adenocarcinoma cell line. DIGE technologies were applied to compare the proteome of A549 (paclitaxel sensitive) and A549-Taxol (paclitaxel resistant) cell lines. By using matrix-assisted laser desorption/ionization time-of-flight tandem mass spectrometry (MALDI-TOF–MS/MS), we identified 30 differentially expressed proteins. The expression of key molecules was verified by Western blot. These proteins may provide a novel clue to elucidate the drug-resistant mechanism of paclitaxel.

Materials and methods

Chemicals

Reagents for 2D-DIGE experiments were bought from GE health care (Uppsala, Sweden). General chemicals were purchased from Sigma–Aldrich (Poole, UK). Bradford assay kit was obtained from Bio-Rad (Hercules, CA, USA). Paclitaxel (PTX) was obtained from Bristol-Myers Squibb (Princeton, USA). Cisplatin (CDDP) was obtained from Merck (Overijse, Belgium). Vinorelbine (NVB) was obtained from Pierre fabre (Lavaur, France). Vincristine (VCR) was obtained from Main Luck Pharmaceuticals (Shenzhen, China). Mitomycin C(MMC) was obtained from New Asiatic Pharmaceuticals (Shanghai, China). Gemcitabine (GEM) was obtained from Eli Lilly company (Indianapolis, USA).

Cell culture and establishment of paclitaxel-resistant A549-Taxol

The paclitaxel-resistant A549-Taxol cell line was established as previously reported. Briefly, the cell lines were selected to be paclitaxel resistant by continuous culture in medium containing stepwise increases in paclitaxel concentration over a period of 8 months. A549-Taxol cells were obtained by biweekly medium changes with chronic exposure to 5 ng/ml paclitaxel, where doses of paclitaxel were escalated stepwise to 1000 ng/ml. During this procedure, a few surviving cells proliferated and recolonized the cultures, yielding a paclitaxel-resistant strain.

Evaluation of A549-Taxol resistance level

To assess the sensitivity of A549-Taxol cell line to various anticancer drugs, a cytotoxicity assay was carried out using 3-[4,5 dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT). In brief, A549 and A549-Taxol cells were harvested, counted, and seeded into 96-well plates at 2 × 103 cell/well. The cells were incubated at 37°C for 24 h prior to the addition of the drugs. Drugs including paclitaxel, cisplatin, vinorelbine, vincristine, mitomycin C, and gemcitabine were diluted to a range of concentrations. They were added to the cells and incubated for 72 h. The cells exposed to culture medium served as an experimental control. Twenty microliters of MTT (5 mg/ml) was added to the cells and incubated at 37°C for 3 h. The MTT solution was removed, and 200 μl of DMSO was added to each well to dissolve the blue formazan crystals. The optical density was measured at 490 nm wavelengths using Model 680 microplate reader (Bio-Rad). IC50 values (the concentration of drugs that produced a 50% reduction of absorbance) of treated and untreated cells were analyzed. To gain accuracy, four wells were used for each condition, and 3 independent MTT assays were repeated.

Immunostaining

Cells were cultured on sterile coverslips inside 6-well plate. Then, the cells were washed with PBS and fixed with precooling acetone for 15 min, Non-specific binding was blocked by incubating the cells with 10% BSA (Sigma) in PBS for 60 min. Slides were then incubated overnight at 4°C with rabbit polyclonal anti-P-glycoprotein (P-gp) (1:100; Santa Cruz Biotechnology) and rabbit polyclonal anti-glutathione S-transferase pi (GSTP1) (1:100; Santa Cruz Biotechnology) for 16 h followed by the secondary biotinylated goat anti-rabbit antibody (1:400; Vector Laboratories). After three washes with PBS, the sections were incubated with horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology) for 1 h at room temperature. Immunocomplexes were detected with 3,3-diaminobenzidine as a chromogen resulting in deposition of a brown reaction product. Samples were counterstained with hematoxylin, the blue nuclear stain. The cells containing brown granule were considered as positive criteria. Five fields of cells (at least 100 cells totally) were counted at a magnification of ×200. The percentage of positive cells and the intensity of staining were assessed in a semi-quantitative manner, and each slide was given a total score based on the results. The percentage of positive cells was recorded by the following criteria: 0–25% for 0; 26–50% for 1; 51–75% for 2; 75% for 3. The intensity of staining was graded: absence or faint yellow for 0; weak for 1; moderate for 2; and strong for 3. Thus, the minimum staining score was 0, and the maximum score was 6.

DIGE

Twenty-four hours after seeding (half-confluence state), cells were washed with deionized water and lysed with 200 μl of lysis buffer containing 30 mM Tris, 7 M urea, 2 M thiourea, and 4% (w/v) CHAPS. Salts were then removed with Ettan 2-D Clean-Up Kit; Protein concentration was adjusted to 5 mg/ml by addition of DIGE labeling buffer. Samples from either A549 or A549-Taxol cells were labeled with Cy3 or Cy5 cyanine dyes while internal standard pooled sample was labeled with Cy2 dye according to manufacturer’s instructions (GE heath care). Samples were finally combined according to the experimental design, at 50 μg of protein per Cydye per gel, and diluted twofold with IEF sample buffer (7 M urea, 4% w/v CHAPS, 130 mM DTT, 1% IPG buffer, 2 mM PMSF). First dimension IEF was performed on IPG strips (13 cm; non-linear gradient pH 3–10) using an Ettan IPGphor system. After focusing at 45 kVh, strips were equilibrated first for 15 min in 6 ml of equilibration solution (6 M urea, 50 mM Tris–Cl, 30% glycerol, 2% SDS) with 10 mg/ml DTT followed by 15 min in equilibration buffer with 25 mg/ml iodoacetamide. Second dimension SDS–PAGE were run by overlaying the strips on 12.5% gels on Hofer SE600 system. Gels were run at 20°C, at constant power 10 mA/gel for 30 min followed by 20 mA/gel until the bromophenol blue tracking front had run off.

Fluorescence images of the gels were acquired on a Typhoon 9400 scanner (GE Healthcare). Cy2, Cy3, and Cy5 images were scanned at 488 nm/520 nm, 532 nm/580 nm, and 633 nm/670 nm excitation/emission wavelengths, respectively, at a 100-μm resolution. Image analysis and statistical quantification of relative protein abundances were performed using DeCyder 2-D Software v.6.5 (GE Healthcare). The differential in-gel analysis module was used for pair-wise comparisons of each A549 and A549-Taxol extracts to the mixed standard present in each gel and for the calculation of normalized spot volume/protein abundance. Replicate gels were used to calculate average abundance changes. Each individual protein was analyzed by using the DeCyder biological variation analysis module. Differential protein spots in resistant and sensitive groups (|ratio| ≥ 1.7, P < 0.01) were marked.

Protein preparation and identification by mass spectrometry

Differentially expressed proteins were identified from preparative gels (Coomassie staining, 500 μg). Protein spots of interest were picked and destained with 5 mM ammonium bicarbonate, 50% ACN. Gels were then dried completely by centrifugal lyophilization. In-gel digestion was performed with 30-ng trypsin (Promega) in 25 mM ammonium bicarbonate for 15 h at 37°C. The supernatants were collected, and the tryptic peptides were extracted from the gel sequentially with 0.2% TFA for 3 times. All extracted solutions were combined and then dried with N2. Peptides were eluted with 50% acetonitrile in 0.1% TFA and mixed at 1:1 with 20 mg/ml α-cyano-4-hydroxy cinnamic acid (CHCA) in acetonitrile before spotting onto a stainless steel MALDI target plate. Spectra were obtained using a mass spectrometer MALDI-TOF–MS/MS (ABI 4700). Trypsin-digested peptides of horse myoglobin were used as mass standard to calibrate the instrument, and then default calibration was applied on the sample peptides. The instrument was used in reflector-positive mode with an acceleration voltage of 20 kV. Four strongest peptides per spot were selected automatically for MS/MS analysis. PMF data were searched with the search engine MASCOT (Matrix Science, London, UK) against Swiss-Prot database.

Western blot analysis

Protein concentrations were determined by the Bradford assay using BSA as standard (Protein Assay Kit, Bio-Rad). Total protein extracts (30 μg) were mixed with SDS sample buffer (62.5 mM Tris–HCl, pH 6.8, 2.3% SDS, 10% glycerol, 5% β-mercaptoethanol, 0.005% bromophenol blue) and resolved by SDS PAGE on 12% acrylamide gels. Proteins were detected immunologically following semidry electrotransfer (Trans-Blot® SD semidry electrotransfer system, Bio-Rad) onto PVDF membranes (Protran, Millipore). The membranes were blocked with 5% non-fat dry milk in PBS for 1 h at room temperature and incubated overnight at 4°C with the following primary antibodies: anti-CK8 (1:500, Santa Cruz, monoclonal) for keratin 8, anti-CK18 (1:500, Santa Cruz, monoclonal) for keratin 18, anti-Calpastatin (1:250, Millipore, monoclonal) for calpastatin, anti-annexin 1 (ANX1) (1:500, Santa Cruz, monoclonal) for annexin 1, anti-RALDH1 (1:200, Santa Cruz, polyclonal) for Retinal dehydrogenase 1. After washing 3 times in Tris-buffered-saline with Tween, blots were incubated with horseradish peroxidase-conjugated secondary antibody (diluted 1:5000, Santa Cruz Biotechnology) for 1 h at room temperature. Immunoreactive complexes were visualized using HRP-DAB Detection Kit (Beijing, Tiagen Biotech).

Statistical analysis

Statistical analysis was performed with SPSS 13.0 software package. Values were presented as the mean ± SEM. Statistical significance was evaluated using the Student’s two-tailed t-test. P < 0.05 was considered statistically significant.

Results

Establishment of paclitaxel-resistant cell lines

A paclitaxel-resistant subline derived from the parental-sensitive cell line A549 was established by stepwise selection in paclitaxel. The subline was designated A549-Taxol. The IC50-value of A549-Taxol against paclitaxel was 5128 ± 0.7 μg/l, compared with 10 ± 0.5 μg/l for A549, a 512-fold resistance. A549-Taxol showed cross-resistance against mitomycin C, vincristine, and Vinorelbine and no resistance against Gemcitabine and cisplatin (Table 1). Comparison between chemosensitive and chemoresistant A549 cells revealed significant differences in the expression of P-gp and GSTP1 (Fig. 1).
Table 1

Cytotoxic effects of different anticancer agents in A549 and A549-Taxol cells

Anticancer agents

IC50a (μg/L)

Relative resistance indexb

A549

A549-Taxol

Paclitaxel

10 ± 0.5

5128 ± 0.7

512.8

MMC

93.7 ± 12.5

6230.8 ± 26.5

66.48

VCR

768.0 ± 24.1

3750.2 ± 23.5

4.88

NVB

30,000.1 ± 100.3

62,000.2 ± 68.5

2.06

aIC50 value is the lethal dosage required inhibiting 50% of cells growth. Every IC50 value obtained is the mean value from three independent MTT assays

bRelative resistance index is defined as the IC50 value of A549-Taxol divided by the IC50 value of A549

https://static-content.springer.com/image/art%3A10.1007%2Fs00432-010-0913-9/MediaObjects/432_2010_913_Fig1_HTML.gif
Fig. 1

Immunocytochemistry for P-gp and GSTP1 of A549 and A549-Taxol cells. Up, representative images (magnification ×200), a P-gp for A549; b P-gp for A549-Taxol; c GSTP1 for A549; d GSTP1 for A549-Taxol. Down, mean staining scores form triplicate cover slides seeded with cells; bars, ± SD. *P < 0.01 versus control, ANOVA

Analysis and identification of differentially expressed proteins

Total extracts from parental and chemoresistant cell lines, together with an internal standard control, were differentially labeled and analyzed by 2D-DIGE. Three replica gels were considered for the quantitative and statistical analysis using the DeCyder™ 6.5 software. This analysis revealed changes in the abundance of 76 proteins, with a statistical variance of the parental versus chemoresistant spot volume ratios within the 99th confidence level (Student’s t-test; P ≤ 0.01). Thirty-five were significantly up-regulated in the chemoresistant cells (ratiochemoresistance/chemosensitive ≥ 1.7, P ≤ 0.01;), whereas 41 were down-regulated (ratiochemoresistance/chemosensitive ≤ −1.7, P ≤ 0.01;). Figure 2 shows a representative 2-D gel image. Arrows indicate those proteins identified whose expression varies within the 99th confidence level.
https://static-content.springer.com/image/art%3A10.1007%2Fs00432-010-0913-9/MediaObjects/432_2010_913_Fig2_HTML.gif
Fig. 2

A representative 2-D map of the total proteins was to be obtained by DIGE analysis. Protein extracts from the A549 and A549-Taxol cells were covalently labeled with Cy3 (green) and Cy5 (red) fluorochromes, respectively. A mixed internal standard combining all the proteins from both extracts, labeled with an additional Cy2 dye was included in all gels. Overlapping image illustrates the changes in protein abundance after treatment of the cells with the drug for a long period of time. Those proteins whose expression varied within statistical significance (Student’s t-test, P < 0.01) and were unequivocally identified by PMF are named and indicated by arrows. Numbers correlated with those were included in Table 2

Figure 3 illustrates the quantitative analysis of four proteins that were chosen for further characterization. A representative 3-D view of the abundance of each protein in two cell lines, as revealed by DIGE analysis of one of the 2-D gel replicas used for the statistical analysis. The graph view shows the average ratio of each protein (referred as the standardized Log abundance, according to the internal standard) in A549 and A549-Taxol.
https://static-content.springer.com/image/art%3A10.1007%2Fs00432-010-0913-9/MediaObjects/432_2010_913_Fig3_HTML.gif
Fig. 3

Comparison analyses of protein spot intensities by DeCyder 6.5. Examples for the evaluation by DeCyder of alterations in spot intensities using the DIGE system are displayed. Spot information of four selected spots, i.e., 666, 784, 1456, and 1487 (according to Table 2), derived from A549 and A549-Taxol cells is shown. To point out alterations in corresponding spot intensity proportions, the selected spots are displayed as three-dimensional (3D) images in the top panel. Three-dimensional images were rotated to allow optimal perception. The bottom panel shows associated graph views of standardized log abundances of the selected spots among analyzed gel replicates. Control stands for A549 cells, Treated stands for A549-Taxol cells

Twenty-seven of 35 up-regulated and 10 of 41 down-regulated protein spots were cut out and analyzed, 30 were unambiguously identified by PMF. The tryptic peptide fingerprint for the spot 666 in Fig. 3 is shown in Fig. 4. These 30 spots corresponded to 21 different proteins. In some cases, the same protein was identified in different spots across the 2-D gel, suggesting the occurrence of post-translational modifications. Table 2 presents these identifications with Swiss-Prot accession numbers, theoretical molecular weights and pI, subcellular location as well as sequence coverage rates. Proteins were grouped according to the biological processes in which they are likely to participate, i.e., signal transduction, cytoskeleton, redox reaction, energy and metabolism, chaperone activity and transport. Most of the proteins showed mitochondrial and cytoplasm location, such as superoxide dismutase [Mn] or aconitate hydratase in the mitochondrial matrix. A few of differential proteins which highly expressed in A549-Taxol showed cytoskeleton location, i.e., CK8 and CK18. Others are integral proteins involved in the transport of molecules across membranes or signal transduction, for example, voltage-dependent anion-selective channel protein 2 (VDAC-2) in the mitochondrial outer membrane.
https://static-content.springer.com/image/art%3A10.1007%2Fs00432-010-0913-9/MediaObjects/432_2010_913_Fig4_HTML.gif
Fig. 4

MALDI-MS analysis of tryptic peptides obtained from spot 666 shown in Fig. 1a. a MALDI-TOF peptide fingerprint. Vertical arrows indicate peptides chosen for MALDITOF/TOF analysis. b Fragmentation map of the 1079.56-Da peptide. c Fragmentation map of the 1129.67 peptide. d Fragmentation map of the 1419.82 peptide

Table 2

Identification of 30 differential proteins between A549-Taxol and A549 cells by MALDI-TOF/MS

Spot No.a

Accession no.b

Protein name

Protein score

Seq. Cov (%)

Theoretical pI/Mr(KD)

Av. ratioc

t-Test

Subcellular location

Heat shock and stress chaperone

 387

P11021

GRP78

198

14

5.07/72.29

−1.98

0.0036

ER

 554

P10809

HSP60

183

19

5.70/61.01

−1.80

0.0032

Mitochondrial

Transcription and translation

 1082

P31942

HnRNP H3

84

10

6.37/36.90

−1.82

0.0019

Nuclear

Cytoskeletal

 647

P05787

CK8

372

41

5.52/53.67

41.40

0.0036

Cytoskeleton

 666

P05787

CK8

290

29

5.52/53.67

36.44

0.001

Cytoskeleton

 778

P05783

CK18

208

28

5.34/48.03

10.04

0.0075

Cytoskeleton

 784

P05783

CK18

208

28

5.34/48.03

44.66

0.0013

Cytoskeleton

 335

P02545

Lamin A/C

148

15

6.57/38.69

3.56

0.01

Nuclear membrane

 336

P02545

Lamin A/C

170

20

6.57/38.69

2.01

0.0097

Nuclear membrane

 679

P68371

Tubulin beta-2C chain

154

40

4.79/49.80

1.84

0.0011

Cytoskeleton

Glycometabolism

 278

P16276

Aconitate hydratase

503

27

7.36/85.37

2.01

0.0061

Mitochondrial

 484

P11413

G6PD

69

13

5.39/59.22

−1.75

0.0047

Cytoplasm

 571

P14618

Pyruvate kinase isozymes M1/M2

228

24

7.96/57.90

−2.07

0.01

Cytoplasm

 835

P00558

Phosphoglycerate kinase 1

125

15

8.30/44.59

−3.52

0.0046

Cytoplasm

 721

P06733

Alpha-enolase

276

29

7.01/47.14

2.26

0.0051

Cytoplasm

 738

P06733

Alpha-enolase

268

32

7.01/47.14

2.22

0.0099

Cytoplasm

Amino acid metabolism

 35

P31327

Carbamoyl-phosphate synthase

375

23

6.30/164.84

18.93

0.0019

Mitochondrial

 38

P31327

Carbamoyl-phosphate synthase

185

12

6.30/164.84

20.64

0.0013

Mitochondrial

 142

P11586

C1-THF synthase

282

27

6.89/101.50

2.19

0.0057

Mitochondrial

Redox reaction

 1454

P09211

GST Class-Pi (GSTP1-1)

235

30

5.43/23.34

2.41

0.0071

Cytoplasm

 1519

P09211

GST Class-Pi (GSTP1-1)

254

22

5.43/23.34

2.12

0.01

Cytoplasm

 1456

P09211

GST Class-Pi (GSTP1-1)

268

31

5.43/23.34

3.40

0.0010

Cytoplasm

 1487

P04179

Superoxide dismutase [Mn]

98

16

8.35/24.70

−4.40

0.001

Mitochondrial

 603

P00352

RALDH1

165

25

5.34/48.03

87.84

0.0012

Cytoplasm

 617

P00352

RALDH1

134

21

5.34/48.03

15.33

0.0016

Cytoplasm

 619

P00352

RALDH1

97

16

5.34/48.03

20.82

0.0012

Cytoplasm

Signal transduction

 304

P15311

Ezrin (p81) (Cytovillin)

83

9

5.94/69.37

3.15

0.0049

Integral membrane

 127

P20810

Calpastatin

79

9

4.98/76.53

2.81

0.0055

Cytoplasm

Transport

 999

P04083

Annexin-1

212

36

6.57/38.69

6.55

0.0015

Integral membrane

 1159

P45880

Voltage-dependent anion-selective channel protein 2 (VDAC-2)

109

24

7.49/31.55

−2.05

0.0012

Mitochondrial outer membrane

GRP78 78 kDa glucose-regulated protein, HSP60 60-kDa heat shock protein, HnRNP H3 heterogeneous nuclear ribonucleoprotein H3, C1-THF synthase C-1-tetrahydrofolate synthase

aSpot numbers correspond to those included in the 2D image (Fig. 2)

bAccess name according to Swiss-Prot

cAverage ratio between A549-Taxol and A549 cell lines calculated considering 3 replica gels

Validation of identified proteins

To validate the identified proteins, we did Western blot analysis to determine the level of these proteins in A549 and A549-Taxol cells. As shown in Fig. 5, CK8, CK18, ALDH1, CAST, and ANX I were up-regulated in the drug-resistant A549-Taxol cells, consistent with the data shown in Table 1 and Fig. 2 obtained using proteomic approach.
https://static-content.springer.com/image/art%3A10.1007%2Fs00432-010-0913-9/MediaObjects/432_2010_913_Fig5_HTML.gif
Fig. 5

Immunodetection of CK8, CK18, CAST, ALDH1, and ANX I by Western blot of A549 and A549-Taxol whole lysates. As suggested by 2-DE, CK8, CK18, CAST, ALDH1, and ANX I appeared overexpressed in A549-Taxol when compared to the sensitive parental cell line. GAPDH was used as the internal control of the protein concentration in the extracts. The relative amounts of differential protein were quantified as ratios to GAPDH. Results represent the average of three independent experiments; bars, ± SD. *P < 0.01 versus control, #P < 0.05 versus control

Discussion

Paclitaxel, originally isolated from Taxul brevifolia (Pacific Yew), is a cytotoxic microtubule-stabilizing agent, consisting of a taxane ring system linked to a rare four-membered oxetane ring and an ester side chain (Parekh and Simpkins 1997). Potential mechanisms of explaining paclitaxel resistance include multidrug resistance, the alterations in β-tubulin, the ability to detoxify paclitaxel, genetic changes like BCL-2, P53, BAD, and other genes involved in apoptosis (Yusuf et al. 2003). However, this process has not yet been fully explained. We aimed to identify key molecules associated with paclitaxel resistance, using high throughput proteomic technologies. Here, we used DIGE to examine differential levels of protein expression prepared from paclitaxel-sensitive A549 and paclitaxel-resistant derivative A549-Taxol cell lines. In the present study, we identified thirty differential proteins, among which five proteins were validated using Western blot.

Previous study showed that some of the proteins overexpressed in paclitaxel-resistant cells were associated with chemoresistance, including RALDH1 (86.44-fold), CK8 (41.40-fold), CK18 (44.66-fold), GSTP1 (3.40-fold) (Le Moguen et al. 2006; Liu et al. 2006; Zhang and Liu 2007). In this study, new proteins involved in chemoresistance similar to calpastatin (2.81-fold) were identified. Potential pathways and roles in drug resistance were discussed as follows.

Drug efflux

Alterations in drug efflux may be mediated by ABC transporter proteins such as P-glycoprotein (P-gp) and multidrug-resistance protein (MRP) (Wilson et al. 2006). These proteins can actively transport drugs out of cells. However, transmembrane proteins such as ABC transporters were not identified in the 2-DE gel, which may be due to their hydrophobic nature, low abundance, and heterogeneity in glycosylation interfering with solubilization and isoelectrofocusing during 2-DE separation. Our results from immunostaining also showed that the chemoresistant cell line overproduced P-gp. There has been conclusive evidences to indicate that paclitaxel is a substrate for P-gp and that overexpression of P-gp can lead to paclitaxel resistance (Sangrajrang and Fellous 2000).

Our study also demonstrated a 6.55-fold increase of annexin 1, an annexin family member. Annexins belong to the ubiquitous family of structurally related proteins that share the common property of reversible Ca2+-dependent binding to membranes containing phosphatidylserine (Zhang and Liu 2007). The mechanism by which annexins confer drug resistance is not known, and drug efflux through exocytosis of drug-filled vesicles may play a role.

Cytoskeleton and paclitaxel resistance

The expression of four cytoskeletal proteins (CK) was increased in paclitaxel-resistant cell line compared to that in chemosensitive cell line, among which CK8 and CK18 showed a higher expression (36.44- and 44.66-fold). CKs are included in the subfamily of intermediate filament proteins. The heterotypic complexes of CK8 and 18 have been implicated in resistance to TNF-α induced apoptosis by binding the cytoplasmic domain of tumor necrosis factor receptor 2 (Caulin et al. 2000). Another study demonstrated that CK8 modulates FAS trafficking from the Golgi to cell surface (Gilbert et al. 2001). Moreover, CK18 could sequester TNFR1-associated death domain (TRADD) to attenuate interactions between TRADD and activated TNFR1, and moderate TNF-induced apoptosis (Inada et al. 2001). All of these results showed that CK8 and 18 might be involved in paclitaxel-induced apoptosis responsible for paclitaxel resistance by modulating trafficking and interactions of some molecules.

Several alterations in the expression patterns of both α and β-tubulin have been identified in paclitaxel-resistant cell lines (Seve and Dumontet 2005; Yusuf et al. 2003). The available data suggest that alterations in microtubule structure and/or function represent an important mechanism of resistance to paclitaxel. Jaffrezou reported that the KPTA5 cell line, which was exclusively resistant to taxanes, displayed the increased expression of IVa tubulin isotype (Jaffrezou et al. 1995). Our study demonstrated a 1.84-fold increase of tubulin beta-2C chain in the A549-Taxol cell line, coinciding with the report of Haber, whose data showed that, in the murine cell line J774, resistance to paclitaxel was associated with a 21-fold increase in class II beta-tubulin isotype (Haber et al. 1995).

Apoptosis

In many tumors, chemoresistance acquisition is due to upregulation or modification of key elements of apoptosis control (Bergman 2003; Sakai 2006). In our research, many proteins involved in signal transduction, such as ezrin and calpastatin, were found differentially expressed in chemoresistant cell line. These results were in accordance with Laura’s report that the aplidin-resistant HeLa cell line displayed increased expression of ezrin (Gonzalez-Santiago et al. 2007). Ezrin is a member of the ezrin/radixin/moesin (ERM) protein complex which involved in cytoskeletal remodeling (Tsukita and Yonemura 1997). By interacting with p85, the regulatory subunit of phoshatidylinositol 3-kinase, ezrin, can activate the protein kinase Akt to protect cells against apoptosis stimulation (Gautreau et al. 1999). Overexpression of ezrin in the A549-Taxol cells could activate the Akt pathway and inhibit the induction of apoptosis by paclitaxel. In addition, ezrin links Fas to the actin cytoskeleton, and it is an essential requirement for the susceptibility to Fas-mediated apoptosis (Fais et al. 2005). Luciani found that P-glycoprotein colocalized and coimmunoprecipitated with ezrin in a multidrug-resistant variant of CCRF-CEM cell, and inhibiting ezrin would reduce drug efflux and induce the redistribution of P-glycoprotein (Luciani et al. 2002).

Calpastatin, which had never been reported with regard to chemoresistance, is a specific endogenous inhibitor of the protease calpain (Emori et al. 1992; Murachi 1990). Calpain is a cytosolic cysteine protease which plays an important role in transducing signals of cell migration, differentiation, and proliferation (Wendt et al. 2004). Calpain also takes part in modulating apoptotic cell death induced by taxol (Impens et al. 2008). More interestingly, ezrin is a substrate for calpain (Gomperts et al. 2004). Calpastatin may reduce apoptosis and play a role in chemoresistance by inhibiting calpain, which needs further research.

Metabolic enzymes

Consistent with our report, a recent study has demonstrated that drug-resistant cancer cells had changed rates of glycolysis. The abundance of two key enzymes, involved in glycolysis, is increased here. Aconitate hydratase and alpha-enolase may maintain the supply of bioenergetic substrates for oxidative metabolism by mitochondria, thus inhibit apoptosis, and result in a resistant phenotype. However, it should be noted that phosphoglycerate kinase 1 and G6PD involved in glycolysis, which have been reported to be up-regulated in chemoresistant cancer cells in other literatures (Murphy et al. 2008), do show decreased abundances in our chemoresistant cell line. The deviation of their expression needs to be explored further.

Moreover, two amino acid metabolic-related proteins were shown to be overexpressed in A549-Taxol, including carbamoyl-phosphate synthase (18.93-fold) and C1-THF synthase (2.19-fold). This suggested that amino acid metabolism might be a very important factor for cancer cells to survive drug treatment.

Reactive oxygen species (ROS) and drug metabolism

Several differentially expressed protein spots were identified as proteins and enzymes involved in different metabolic pathways. Aldehyde dehydrogenase (ALDH) showed a dramatically increased expression in paclitaxel-resistant cell lines (86.44-fold). It has been demonstrated, in several cell lines, that paclitaxel can induce apoptosis in a reactive oxygen species (ROS)-dependent manner (Gervasoni et al. 2004; Wang et al. 2004), with toxic effects of producting toxic aldehydes by lipid peroxidation. ALDH is implicated in the detoxification of these aldehydes as well as in the sensitization of certain cells to aldehydes (Canuto et al. 1994). ALDH maybe could lead to paclitaxel resistance by decreasing the production of ROS and ROS-dependent apoptosis of paclitaxel.

Another representative protein we detected was Glutathione S-transferase P1 (GSTP1). This protein was highly expressed in chemoresistant cells and identified in three different spots across the 2D gel. The abundant isoform of glutathione S-transferases (GSTs) in lung epithelium plays a key role in cellular protection against oxidative stress and toxic foreign chemicals (Board et al. 1998). It had been reported that GSTP1 affects resistance to camptothecin in human lung adenocarcinoma cells and the increased expression of the GSTP1 subgroup correlated with resistance to cisplatin in ovarian cancer cells and tumors (Ishii et al. 2004; Li et al. 2004).

In conclusion, our work provides a profile of the differential protein expression in lung adenocarcinoma cells associated with paclitaxel resistance using DIGE/MS. Some of these alterations may provide a major evasion route for the cells to escape the induction of apoptosis by paclitaxel and help to design customized models. In addition, we plan to analyze fresh tumor samples (resistant vs. sensitive) from patients in a similar manner to verify our findings. Our work may be useful to elucidate chemoresistance mechanisms, so as to develop tailored therapeutic strategies.

Acknowledgments

This work was supported by a grant from the Shanghai United Municipal Hospitals Project (to Hui-fang SHA) (NO.SHDC12007103), Shanghai Municipal Natural Science Foundation (to Qiang-ling SUN) (NO.10ZR1428100), and grant for cancer research from Shanghai charity foundation (to Qiang-ling SUN). We wish to thank all the researchers of public service platform of Shanghai Shenkang medical center which located in Ruijin Hospital for their assistance with the mass spectrometric analyses in this study.

Conflict of interest statement

None.

Copyright information

© Springer-Verlag 2010