Article pubs.acs.org/jpr
Quantitative Proteomics Reveals That the Inhibition of Na+/ K+‑ATPase Activity Affects S‑Phase Progression Leading to a Chromosome Segregation Disorder by Attenuating the Aurora A Function in Hepatocellular Carcinoma Cells Zhongwei Xu,† Fengmei Wang,‡ Fengxu Fan,† Yanjun Gu,§ Nana Shan,† Xiangyan Meng,∥ Shixiang Cheng,§ Yingfu Liu,† Chengyan Wang,† Yueying Song,† and Ruicheng Xu*,⊥ †
Central Laboratory, Logistics University of Chinese People’s Armed Police Force, Tianjin 300162, China Department of Gastroenterology and Hepatology, The Third Central Hospital of Tianjin, Tianjin 300170, China § Affiliated Hospital of Chinese People’s Armed Police Force, Tianjin 300162, China ∥ Department of Physiology and Pathophysiology, Logistics University of Chinese People’s Armed Police Force, Tianjin 300162, China ⊥ Tianjin Key Laboratory for Biomarkers of Occupational and Environmental Hazard, No. 1 Huizhi Huan Road, DongLi District, Tianjin 300309, China ‡
S Supporting Information *
ABSTRACT: Many studies have shown the Na+/K+-ATPase (NKA) might be a potential target for anticancer therapy. Cardiac glycosides (CGs), as a family of naturally compounds, inhibited the NKA activity. The present study investigates the antitumor effect of ouabain and elucidates the pharmacological mechanisms of CG activity in liver cancer HepG2 cell using SILAC coupled to LC−MS/MS method. Bioinformatics analysis of 330 proteins that were changed in cells under treatment with 0.5 μmol/L ouabain showed that the biological processes are associated with an acute inflammatory response, cell cycle, oxidation reduction, chromosome segregation, and DNA metabolism. We confirmed that ouabain induced chromosome segregation disorder and S-cell cycle block by decreasing the expression of AURKA, SMC2, Cyclin D, and p-CDK1 as well as increasing the expression of p53. We found that the overexpression or inhibition of AURKA significantly reduced or enhanced the ouabain-mediated the anticancer effects. Our findings suggest that AURKA is involved in the anticancer mechanisms of ouabain in HepG2 cells. KEYWORDS: Na+/K+-ATPase, ouabain, AURKA, SILAC, proteomics, hepatocellular carcinoma
■
INTRODUCTION Na /K+-ATPase (NKA) is a kind of universally expressed transmembrane transporter of tetramer alpha and beta subunits.1 The cardiac glycosides (CGs), which are wellestablished cellular targets for NKA, induce the sodium gradient and, consequently, the activity of the Na+/Ca2+ exchanger, leading to an increase in intracellular calcium ion concentration and enhancing the myocardial contractility. Thus, they are extensively applied for the treatment of heart failure;2 however, many studies have revealed that very few patients that were treated with CGs for cardiovascular diseases died from cancer, suggesting CG’s possible application to the oncology.3 CGs include compounds obtained not only from plants (ouabain, digoxin, digitoxin, and oleandrin) but also from amphibians (bufalin and marinobufagenin).4 Several studies provide evidence that NKA is important for the progression of the epithelial-to-mesenchymal transition in cancer cells.5−7 Moreover, many patients fail to respond to
cancer chemotherapy due to the multi-drug-resistant (MDR) phenotype during chronic treatment.8 Therefore, the NKA may be a critical target in MDR cancer cells. The NKAs that are localized in caveolae cause conformational variation to neighboring proteins, which activates the multiple signaling pathways, including the Src-dependent transactivation of EGFR9,10 and the activation of PI3K/Akt,11 which may result in the activation of the ERK1/2 signaling pathway,12 and increases the ROS production and then activates JNK pathway.13,14 Hepatocellular carcinoma (HCC) is a high mortality disease.15 Our previous studies showed that the NKA of liver cancer tissues or cells exhibited a higher expression level and provided the basis for CG selectivity against malignant cells.16 Two kinds of CGs, Anvirzel and UNBS1450, have entered
+
© XXXX American Chemical Society
Received: May 15, 2015
A
DOI: 10.1021/acs.jproteome.5b00724 J. Proteome Res. XXXX, XXX, XXX−XXX
Article
Journal of Proteome Research
Figure 1. Quantitative proteomics comparison of the HepG2 cells treated with ouabain using the SILAC approach. (A) Work flow for the quantitative comparison of the change in amount of protein as a function of ouabain treatment. (B) Histograms of log ratios of abundance of quantified proteins. (B1) Forward labeled (n = 3734) and (B2) reverse labeled (n = 3955) experiments. (C) Global distribution of the overlap of the qualified proteins in the swap-labeled SILAC samples. (C1) Proteins quantified in both experiments (n = 3193) are shown in the central scatter plot. (C2) Distribution of 327 differentially expressed proteins in the swap-labeled SILAC samples. Of these, 75 proteins exhibited down-regulated expression and 255 proteins exhibited up-regulated expression.
Cell Culture and SILAC Labeling Process
clinical trial stage, and they are exhibiting the stronger anticancer properties than platina derivatives, such as nedaplatin, oxalipatin, and cisplatin; however, this research lacks a comprehensive understanding mechanism for these drugs.17,18 The present study analyzed the signaling pathways of the inhibition of the NKA activity in the HepG2 cells, upon treatment with ouabain for 24 h using a SILAC-based quantitative proteomics strategy. The inhibition of the NKA by ouabain was found to possibly affect the mitotic cell cycle, the DNA repair process, and the spindle formation process. The down-regulation of the AURKA expression suppressed the chromosome segregation and induced the cell cycle S-phase block and DNA repair. The results clarify the possible anticancer mechanism of the ouabain inhibition of the NKA activity in a HCC.
■
HepG2 cells was from ATCC (Manassas, VA), and they were SILAC-labeled in a H-DMEM medium (-Lys) containing a 10% FBS that was supplemented with 90 mg/mL of 13C6 Llysine or the corresponding nonlabeled amino acid. For the SILAC labeling, HepG2 was cultured in either light or heavy SILAC-labeling medium for 14 days to achieve complete SILAC incorporation. The media was renewed daily, the cells were passaged every 3 to 4 days, and then they were completely labeled. The HepG2 cells were seeded at 5 × 103/well in a 96well microplate containing culture media with 10% FBS, and they were then treated with 0.001, 0.01, 0.1, 1, and 10 μmol/L ouabain for 24, 48, and 72 h. The MTT assay was performed as previously described.19 Each sample was assayed in triplicate. SILAC and Proteomics Assay
A quantitative high-throughput proteomic analysis using SILAC was performed to investigate the potential changes in the HepG2 cell proteome in response to the ouabain-inhibited NKA activity (Figure 1A). In the forward experiments, the corresponding medium that was heavily labeled with HepG2 cells was cultured in 0.5 μmol/L ouabain for 24 h, and the lightly labeled untreated control group was incubated with the renewed medium for 24 h. In the reverse experiments, the labels of the previously mentioned “heavy” and “light” conditions were exchanged; however, the medium labeled circumstance was not changed. The cells were lysed in buffer (8 Murea, 50 mM NH4HCO3, 10 mM DTT, 50 mM IAA, cocktails). Protein concentrations were measured using BCA method. For improving the accuracy of analysis, we applied an exchange labeling strategy, in which a pair of experiments forward experiment [LightHepG2 (control) vs HeavyHepG2 (ouabain)] and reverse experiment [HeavyHepG2 (control) vs Light HepG2 (ouabain)] was performed. Equal amounts of the
METHODS
Antibodies and Reagents
Ouabain was from Sigma-Aldrich. H-DMEM that was deficient in lysine and the dialyzed FBS were both obtained from Thermo-Fisher (New York). The 13C6-lysine (L-lysine) (99% purity) was from Cambridge Isotope Laboratories (Andover, MA). The rabbit monoclonal anti-AURKA (ab52973), SMC2 (ab10412), Cyclin D (ab134175), p-CDK1 (phospho Y15, ab133463), p53 (ab32049), antiactive/pro Caspase3 (ab47131), and the Aurora A inhibitor, PF-03814735, were all purchased from Abcam. The rabbit polyclonal anti-Mre11 (sc-22767), Nbs1 (sc-11431), and RAD50 (sc-20155) were obtained from Santcruz. The antirabbit IgG, the HRP, and the TRITC-labeled secondary antibodies were from KPL (Gaithersburg, MD). B
DOI: 10.1021/acs.jproteome.5b00724 J. Proteome Res. XXXX, XXX, XXX−XXX
Article
Journal of Proteome Research lysates from each of the light- (200 μg) and heavy-labeled samples (200 μg) were mixed, and then they were digested in solution with 12.5 ng/μL of Lys C overnight, as previously described. They were then dried using vacuum centrifugation and stored at −80 °C. The peptide mixture in every SILAC was loaded onto a polySULFOETHYL A strong cation exchange column (250 mm × 4.6 mm i.d., 5 μm, Poly LC) to divide the mixture into 20 fractions. The peptides were analyzed by LC− MS/MS on an LTQ-OrbitrapVelos mass spectrometer (Thermo Electron, CA). Survey scan was operated in the orbitrap analyzer with a resolution of 30 000 at m/z 400. The mass scan range was set between m/z 300−1600. For MS1, the AGC (auto gain control) was set at 1 × 106, and the maximum ion injection time was 150 ms. For MS2, the 20 strongest most intense ions with charge ≥2 were selected to fragmentation via CID mode with 35% normalized collision energy. The mass window for precursor ion selection was m/z 2. For each scan, 5000 ions accumulation was performed in a maximum injection time of 25 ms. The dynamic exclusion range was 30s. The raw data acquired were searched using the SorcererSEQUEST (version 4.0.4 build) against the NCBI human reference protein database (version 02032012, 35 814 protein sequences) supplemented with a pseudoreverse decoy database. The pseudoreverse database was created by reversing all possible tryptic peptides. Search parameters included dynamic modifications for methionine oxidation (+15.99492) and SILAC 13C6-Lys (+6.020129) and static modification for cysteine carbamidomethylation (57.021465). Enzyme specificity was semi-LysC with two missed cleavages allowed. Parent ion mass tolerance was set to 20 ppm, and fragment ion tolerance was 0.5 Da. Peptides with fewer than seven amino acids were excluded and the peptide matches were filtered to a false discovery rate (FDR) of
