TY - JOUR
AU - Ching, Chi B.
AB - Abstract Because of their attractive chemical and physical properties, graphitic nanomaterials and their derivatives have gained tremendous interest for applications in electronics, materials, and biomedical areas. However, few detailed studies have been performed to evaluate the potential cytotoxicity of these nanomaterials on living systems at the molecular level. In the present study, our group exploited the isobaric tagged relative and absolute quantification (iTRAQ)–coupled two-dimensional liquid chromatography-tandem mass spectrometry (2D LC-MS/MS) approach with the purpose of characterizing the cellular functions in response to these nanomaterials at the proteome level. Specifically, the human hepatoma HepG2 cells were selected as the in vitro model to study the potential cytotoxicity of oxidized single-walled carbon nanotubes (SWCNTs) and graphene oxide (GO) on the vital organ of liver. Overall, 30 differentially expressed proteins involved in metabolic pathway, redox regulation, cytoskeleton formation, and cell growth were identified. Based on the protein profile, we found oxidized SWCNTs induced oxidative stress and interfered with intracellular metabolic routes, protein synthesis, and cytoskeletal systems. Further functional assays confirmed that oxidized SWCNTs triggered elevated level of reactive oxygen species (ROS), perturbed the cell cycle, and resulted in a significant increase in the proportion of apoptotic cells. However, only moderate variation of protein levels for the cells treated with GO was observed and functional assays further confirmed that GO was less cytotoxic in comparison to oxidized SWCNTs. These finding suggested that GO was more biocompatible and could be a promising candidate for bio-related applications. cytotoxicity, oxidized single-walled carbon nanotubes, graphene oxide, proteome analysis, oxidative stress, apoptosis Among the various types of nanomaterials, carbon nanotubes (CNTs) and graphene have attracted increasing attention due to their mechanical, optical, electrical, thermal, and structural properties. Recently, the functionalized derivatives of these nanomaterials have been extensively studied for bio-related applications, such as drug delivery, tissue engineering, biomaterials, biosensor, and bioimaging (Choi et al., 2010; Hu et al., 2010; Liu et al., 2008, 2010; Robinson et al., 2008; Sun et al., 2008; Veetil and Ye, 2009). With the growing potential of bio-related applications of these nanomaterials and the mounting societal concern about nanosafety, it has become imperative to evaluate the potential risks that might be caused by these nanomaterials and have a detailed mechanistic understanding of the interactions between these nanomaterials and living systems (Chou et al., 2008; Ding et al., 2005; Manna et al., 2005; Mu et al., 2009; Zhang et al., 2006; Zhou et al., 2008). Previous investigations have demonstrated that single-walled carbon nanotubes (SWCNTs) induce a noticeable cytotoxicity to mammalian cells (Ding et al., 2005; Magrez et al., 2006; Manna et al., 2005; Pulskamp et al., 2007; Raja et al., 2007; Shvedova et al., 2003). More recently, graphene has also been reported to induce a certain degree of cellular toxicity (Hu et al., 2010; Ryoo et al., 2010). Among the suggested toxicity mechanisms, oxidative stress and mechanical rupture of cell membranes have been considered the most acceptable mechanisms to explain the cytotoxicity introduced by these nanomaterials. However, for the functionalized graphitic nanomaterials that have great potential for bio-related applications, such information is generally lacking to date. Comparative proteome analysis has been widely used to study cellular functions in response to various external stimuli. The change of protein profile directly reflects the biological processes the cell undergoes. Thus, it will be of great interest to obtain the protein profile change of living systems caused by nanomaterials and have a fundamental understanding of the interactions between these nanomaterials and living systems at the proteome level. In the present study, we applied the isobaric tagged relative and absolute quantification (iTRAQ)–based comparative proteomics technique to simultaneously detect and quantify the differences of protein expression levels in living systems after the treatment of functionalized graphitic nanomaterials. Briefly, the iTRAQ technique is a liquid chromatography (LC)–based approach, which includes LC separation of isotopically labeled peptides followed by peptide identification and quantification using MS/MS analysis (Ross et al., 2004). In this study, two commonly used derivatives of graphitic nanomaterials, namely, oxidized SWCNTs and graphene oxide (GO), were selected as the testing materials and human hepatoma HepG2 cells were chosen as the in vitro model for studying the potential cytotoxicity on the liver. From our comparative protein profile obtained here, 30 proteins with significant change of expression level were identified from either the oxidized SWCNTs or GO-treated cells. These included metabolic enzymes, signaling proteins, cytoskeletal proteins, and those proteins involved in cell growth and redox regulation of the cell. Based on the protein profile, oxidized SWCNTs induced oxidative stress and perturbed intracellular metabolic routes, protein synthesis, and cytoskeletal systems. Further experiments, such as reactive oxygen species (ROS) measurement and apoptosis detection, confirmed that oxidized SWCNTs did generate an elevated level of ROS and lead to a significant increase in the proportion of apoptotic cells. Cell cycle analysis showed that cells were arrested at G2/M phase after exposure to oxidized SWCNTs. In contrast, only moderate variation of protein levels for the cells treated with GO was observed. Functional assays further confirmed that there was no significant increased percentage of apoptotic cells, and cell cycle was not severely perturbed. Our data indicated that GO was more biocompatible and could be a promising candidate for bio-related applications. The possible mode of action for these functionalized graphitic nanomaterials that led to distinct pattern of protein profile was also discussed. Our findings provided molecular evidence to the cellular functions in response to oxidized SWCNTs and GO, which should be of great importance in evaluating the biocompatibility of these nanomaterials. MATERIALS AND METHODS Preparation and characterization of oxidized SWCNTs and GO. Purified HiPco SWCNTs (Unidym) produced by chemical vapor deposition were used in the present study. These SWCNTs appeared to have a Gaussian distribution of diameters, with a maximum peak around 0.8–1.2 nm. A wet chemical technique that has been tested effective at removing both carbonaceous and catalytic impurities was applied to further purify SWCNTs (Porter et al., 2009) before being used to generate functionalized SWCNTs. Oxidized SWCNTs were produced by refluxing in a mixture of concentrated sulfuric and nitric acids at 70°C for 4 h, as described in detail elsewhere (Zhang et al., 2007). Following the oxidizing process, the oxidized SWCNTs were extensively washed with distilled water (pH adjusted to 7) to remove any remaining impurities. The GO was synthesized from synthetic graphite powder (Sigma-Aldrich) based on a modified Hummers method (Kovtyukhova, 1999). The exfoliation of GO was achieved by ultrasonication of the dispersion for 2 h. Any remaining chemicals or residuals were removed via dialysis with distilled water, pH adjusted to 7. Both the quality of oxidized SWCNTs and GO were checked using atomic force microscopy (AFM), Fourier transformed infrared (FTIR) spectroscopy, and energy-dispersive X-ray spectroscopy (EDS). The AFM was carried on a MFP3D microscope (Asylum Research) with a silicon cantilever operating in tapping mode. FTIR spectra were obtained on a Perkin Elmer Spectrum One Spectrometer. EDS was obtained using JEOL scanning electron microscope (JSM-6390), equipped to perform elemental chemical analysis. Cell cultures. Human hepatoma HepG2 cells were obtained from the American Type Culture Collection (ATCC). The cells were cultured in Dulbecco’s Modified Eagle’s Medium (supplemented with 10% fetal bovine serum, 100 units/ml penicillin, and 100 units/ml streptomycin) and maintained at 37°C in an atmosphere of 5% CO2. All culture media and media supplements were obtained from Life Technologies. Both oxidized SWCNTs and GO were ultrasonicated 1 h for dispersion before being added to the culture media. The readily water soluble oxidized SWCNTs and GO were then diluted to 1 μg/ml in DMEM medium supplemented with 10% FBS and mixed thoroughly. These nanomaterial-supplemented culture media appeared to be stable under the subsequent experimental conditions. After reaching 80% confluence, the cells were treated with nanomaterials by replacing normal media with oxidized SWCNTs or GO-supplemented media, respectively. For the control cells, the fresh media without any nanomaterial supplementation were used. Cell lysis, protein digestion. and labeling with iTRAQ reagent. After exposure to oxidized SWCNTs or GO for 48 h, the cells were collected and lysed in 300 μl of 8M urea, 4% (wt/vol) 3-[(3-cholamidopropyl)dimethylammonio]-2-hydroxy-1-propanesulfonate, and 0.05% (wt/vol) sodium dodecyl sulphate (SDS) on ice for 30 min with regular vortexing. Samples were centrifuged at 16,100 × g for 60 min at 4°C, supernatant was collected, and the protein concentrations were measured by the Quick Start Bradford protein assay (Bio-Rad). A standard curve was made using bovine serum albumin as a control. A total of 100 μg of each sample was further purified by 2-D clean up kit (GE Healthcare) before being used to generate iTRAQ-labeled peptides. The iTRAQ labeling process was carried out according to the manufacturer’s protocol (iTRAQ Reagents Multiplex kit; Applied Biosystems, Foster City, CA). Specifically, the purified proteins were first resuspended in 20 μl of triethylammonium bicarbonate buffer supplemented with 0.1% SDS. To fully dissolve proteins, the sample tubes were put into a sonicator bath for an additional 1 min. Then the samples were subjected to reduction by dithiothreitol at 60°C for 1 h, followed by alkylation using iodoacetamide at room temperature for 15 min. Next, the cysteine-blocked protein samples were digested to peptides using 10 μl of 0.5 μg/μl sequencing-grade modified trypsin (Promega, Madison, WI) at 37°C overnight. The digested samples were labeled with the iTRAQ tags as follows: control HepG2 cells, iTRAQ 114; HepG2 cells incubated with 1 μg/ml oxidized SWCNTs, iTRAQ 115; HepG2 cells incubated with 1 μg/ml GO, iTRAQ 116. After incubation at room temperature for 1 h, the labeled samples were pooled, vacuum dried, and resuspended in 300 μl ddH2O containing 0.1% formic acid and 5% acetonitrile, to achieve approximately 1 μg/μl peptide mixture. The resuspended sample was filtered to remove bulky particles before loading on liquid chromatography/tandem mass spectrometry (LC-MS/MS) for analysis. Online 2D LC-MS/MS analysis. The LC-MS/MS system was equipped with Agilent 1200 Series liquid chromatograph and 6530 Quadrupole-Time of Flight (Q-TOF) mass spectrometer (Agilent Technologies, Santa Clara, CA). The LC system for separation of peptides included a PolySulfoethyl A strong cation exchange (SCX) column (0.32 × 50 mm, 5 μm particles; PolyLC, Inc.) and HPLC-Chip (Large capacity chip II; Agilent Technologies). Specifically, the HPLC-Chip was incorporated a 160-nl enrichment column and a 150 mm × 75 μm reverse-phase analytical column that was packed with Zorbax 300SB-C18 5-μm particles. In the first dimension, a total of 2 μg of the reconstituted peptide mixture was loaded onto the PolySulfoethyl A SCX column. Nine different molar concentrations of 20, 40, 60, 80, 100, 150, 300, 500, and 1000mM ammonium formate solution were used to stepwise elute peptides from SCX column. Notably, the pH of salt solution was adjusted below 3 using formic acid in order to achieve the desired fraction result. The sequentially eluted peptides was first trapped onto the Zorbax 300SB-C18 enrichment column during the enrichment mode of HPLC-Chip and washed isocratically with buffer containing 5% acetonitrile and 0.1% formic acid at 0.004 ml/min for an additional 30 min in order to remove any excess reagent. Next, the HPLC-Chip was changed to analysis mode and the peptides binding to enrichment column was subjected for further separation in the second dimension. Briefly, the enrichment column was switched from the capillary flow path connected with SCX column to the nanoflow pathway followed by an analytical column. The peptides trapped onto the enrichment column were eluted with buffer A (0.1% formic acid) and buffer B (90% acetonitrile + 0.1% formic acid) at a flow rate of 300 nl/min. A linear gradient with two segments was used. One segment was from 5 to 45% buffer B during the initial 40 min, followed by a second segment with a sharp increase to 90% buffer B over 5 min. The flow was held for an additional 5 min at 90% buffer B to elute any remaining substances that might bind to the columns and then allowed for reconditioning to the initial state for 10 min. Further separation was achieved onto the analytical Zorbax 300SB C18 reversed-phase column (75 μm × 150 mm, 5-μm particles). The column effluent was directly analyzed by the 6530 Q-TOF mass spectrometer interfaced through an HPLC-Chip Cube nanospray source. Overall, 10 runs with different fraction of peptides eluted from SCX column were performed to complete one experiment. For the mass spectrometer, the capillary voltage was set to 1800 V to achieve good and stable spray of effluent. The MS data were acquired in the positive ionization mode using Agilent MassHunter Workstation Q-TOF B.02.01. The fragmentor voltage, skimmer voltage, and octopole RF were set to 175, 65, and 750 V during the data acquisition process. Auto-MS/MS with a total cycle time of 1.225 s was chosen for MS/MS acquisition. Specifically, MS spectra were acquired at 8 Hz (eight spectra/s) (m/z 100–2000) in each cycle, and the four most abundant ions (with charge states 2+ and 3+) exceeding 1000 counts were selected for MS/MS analysis at 4 Hz (four spectra/s) (m/z 50–2000). A medium isolation (4 m/z) window was used for precursor isolation. Collision energy with slope of 3.6 V/100 Da and offset of 4 V was used for the fragmentation of selected precursor peptide ions, which was generally enough to achieve iTRAQ reporter ions, were among the most intense ions in the spectra. Reference mass correction was activated using reference mass of 121 and 922. Precursors were set in an exclusion list for 0.5 min after two MS/MS spectra. Data analysis and interpretation. Peptide and protein identifications were carried out by the Spectrum Mill Proteomics Workbench (Rev A.03.03.084 SR4) software from Agilent Technologies. Default settings on Spectrum Mill data extractor program were used to generate the peak list. Peptides were automatically identified by the Spectrum Mill Proteomics Workbench software using UniProtKB/Swiss-Prot protein database (Geneva, Switzerland) search for species of Homo sapiens. The search criteria were set using the following parameters: a mass tolerance of ± 2.5 Da for the precursor ions and a tolerance of ± 0.7 Da for the fragment product ions were used, two missed cleavages were allowed, methylmethanethiosulfate-labeled cysteine and iTRAQ modification at the amino group were set as fixed modification. Autovalidation was carried out to search the proteins and these criteria provided high confidence for the searched proteins that always represent valid results (above 99% confidence). To assure a fair comparison of the results, the same criteria were chosen to the analysis of other independent batches. Relative quantification of peptide was obtained from the MS/MS spectra and was the ratio of the peak intensity at 114.1, 115.1, and 116.1 Da that correspond to the iTRAQ reporter ions. The relative abundance of specific protein was calculated using the relative ratios of all the peptides originating from the same protein. To account for small variation during the sample preparation process, the relative ratios of proteins were normalized based on the overall ratios of the iTRAQ reporter ions from the MS/MS spectra. In the present study, only proteins with two or more qualified peptide matches and relative ratio of > 1.2 or < 0.8 were subjected to further analysis. Cell proliferation assay and intracellular ROS measurement. The cell proliferation was evaluated by the colorimetric 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma) test. Briefly, following HepG2 cells incubation with oxidized SWCNTs or GO for 48 h, the media were aspirated and replaced with 90 μl of serum-free media. Then 10 μl of MTT stock solution (5 mg/ml) was added to each well, followed by incubation for 4 h at 37°C. The supernatant was then removed, and cells were lysed with 100 μl of dimethyl sulfoxide and the absorbance was recorded at 550 nm on a microplate reader (Benchmark Plus). The intracellular ROS level was measured by the oxidation of the nonfluorescent probe dihydrofluorescein diacetate (DFDA) to the fluorescent metabolite fluorescein as reported earlier (Hempel et al., 1999). Briefly, 1 × 105 cells were seeded in each well of 96-well plates and allowed to attach overnight. After incubation of 1 μg/ml of oxidized SWCNTs and GO for 12 h, the media were aspirated and replaced with 100 μl of 5μM DFDA in phosphate-buffered saline (PBS). Cells were incubated with protection from light for 1 h at room temperature. For the quantification of fluorescein generated, the fluorescence intensities were acquired at 488 nm excitation and 535 nm emission with a fluorescence spectrophotometer. Specifically, the fluorescence intensities at the initial loading of probe and after 1 h incubation were both recorded. The absolute intensity of fluorescein formed by the cell was obtained by deducting the background fluorescence intensity at the initial loading of probe. Apoptosis assay and cell cycle analysis. To detect apoptosis at the early stage, Annexin V-fluorescein isothiocyanate (FITC) Apoptosis Detection kit from BD Pharmingen was used. Briefly, cells were seeded in each well of 12-well plates and allowed to attach overnight. After treatment with oxidized SWCNTs or GO for 48 h, cells were collected after trypsin treatment. After washing twice with PBS, 1 × 105 cells were then incubated with annexin V-FITC and propidium iodide (PI) in the 100 μl binding buffer for 15 min at room temperature with protection from light. The cells were diluted fivefold using binding buffer before being subjected to flow cytometry analysis (Beckman Coulter Altra). For cell cycle analysis, 1 × 106 cells were collected after trypsin treatment and fixed with 10 ml of 70% ethanol at 4°C overnight. After repeated washes with PBS, cells were then stained with staining solution (0.1% Triton X-100, 10 μg/ml PI, and 100 μg/ml RNase A in PBS) for 10 min at 37°C. The percentage of cells in different phases of the cell cycle was measured by flow cytometric analysis. Statistical analysis. To gain statistical evidence for the differentially expressed proteins, three independent experiments were performed and data were displayed as mean ± SD. Student's t-test was employed to test the deviation between different batches. Values were only considered to be statistically significant when p values were less than 0.05. RESULTS Characterization of Oxidized SWCNTs and GO As can be seen in Figure 1A, the AFM results revealed that the majority of these white streaks of oxidized SWCNTs appeared to be around 1000–2000 nm. From the height profile, we found the oxidized SWCNTs had diameters in the range of 1.0–6.0 nm, which might correspond to the expected height of individual and small bundles of oxidized SWCNTs. The resulting oxidized SWCNTs were readily soluble and stable in ddH2O, which indicated that the successful functionalization of SWCNTs as the pristine SWCNTs were more likely to form large bundles and precipitate in ddH2O without the supplementation of surfactants. The AFM image revealed that the resulting GO with lateral dimensions of 100 nm was observed (Fig. 1B). From the height profile, the GO nanosheets were single layered with a mean height of ∼1.0 nm, which was thicker than the graphene sheet due to the presence of covalently bound oxygen functional groups and the displacements of the sp3 hybridized carbon atoms (Wu et al., 2009). FIG. 1. View largeDownload slide Tapping-mode AFM image of oxidized SWCNTs (A) and GO (B). FIG. 1. View largeDownload slide Tapping-mode AFM image of oxidized SWCNTs (A) and GO (B). Next, FTIR spectroscopy was used to study the oxidation of SWCNTs after strong acid treatment. As can be seen in Figure 2A, FTIR results supported the theory of nanotube functionalization. From the spectrum, we observed multiple strong adsorption bands for the treated SWCNTs but not for the purified SWCNTs. Because SWCNTs themselves are not Raman active, any infrared-active mode for the treated SWCNTs should be associated with the attached functional groups. From the FTIR result shown in Figure 2B, the GO spectrum showed strong bands corresponding to the C=O stretching vibration of the COOH groups at 1730/cm, the O-H deformations of the C-OH groups at 1370/cm, the epoxide groups at 1220/cm, and the C-O stretching vibration at 1050/cm (Chen and Yan, 2010). In contrast, there were no such bands for graphene as shown from the spectrum. Moreover, EDS result confirmed that metal catalysts that might produce reactive radicals and generate ROS were completed removed for both oxidized SWCNTs and GO (see Supplementary Data). FIG. 2. View largeDownload slide FTIR spectra of oxidized SWCNTs in comparison to purified SWCNTs (A) and GO in comparison to graphene (B). FIG. 2. View largeDownload slide FTIR spectra of oxidized SWCNTs in comparison to purified SWCNTs (A) and GO in comparison to graphene (B). FIG. 3. View largeDownload slide (A) Representative chromatogram result of peptides eluted from the fraction of 100mM ammonium formate. (B) Representative MS window showing four peptide precursors as marked by a red diamond were detected at 15.8 min. (C) Representative MS/MS spectrum of precursor ion with 448.3 m/z. FIG. 3. View largeDownload slide (A) Representative chromatogram result of peptides eluted from the fraction of 100mM ammonium formate. (B) Representative MS window showing four peptide precursors as marked by a red diamond were detected at 15.8 min. (C) Representative MS/MS spectrum of precursor ion with 448.3 m/z. Online 2D LC-MS/MS Analysis Results To investigate how the protein profile changed in response to nanomaterials treatment, the global proteins were extracted from the cells and analyzed by iTRAQ-coupled two-dimensional (2D) LC-MS/MS approach. In this approach, an SCX column and HPLC-Chip (Large capacity chip II; Agilent Technologies) were used to set up the working platform. In the first dimension, the sample peptides were all loaded on the SCX columns, eluted by stepwise injections of salt plugs. The eluted peptides were trapped on the short enrichment column during the enrichment mode. Subsequently, the enrichment column was switched to analytic mode and the peptides trapped on enrichment column were eluted by a nanoflow gradient of acetonitrile and further separated on the reverse-phase column incorporated in HPLC-Chip. As shown in Figure 3A, a representative chromatogram result showed peptide signals for the fraction eluted by 100mM ammonium formate were detected around 8–35 min. A representative MS window showed four precursor ions as marked by a red diamond were detected at 15.8 min (Fig. 3B). These ions met the searching criteria during data acquisition process were selected for further electron bombardment. As can be seen from the MS/MS spectrum shown in Figure 3C, a precursor ion with 448.3 m/z was fragmented for peptide identification and quantification. The peptide identity was derived from the fragmentation pattern of precursor ion and the relative ratio of peptides was calculated from the relative ratio of iTRAQ reporter ions (see Supplementary Data). FIG. 4. View largeDownload slide Differentially expressed proteins involved in redox regulation of the HepG2 cells after treated with oxidized SWCNTs and GO. CNT:C is the ratio of protein expression level in the cells treated with oxidized SWCNTs relative to the control cells; GO:C is the ratio of protein expression level in the cells treated with GO relative to the control cells. Data are shown as mean ± SD from three independent experiments. FIG. 4. View largeDownload slide Differentially expressed proteins involved in redox regulation of the HepG2 cells after treated with oxidized SWCNTs and GO. CNT:C is the ratio of protein expression level in the cells treated with oxidized SWCNTs relative to the control cells; GO:C is the ratio of protein expression level in the cells treated with GO relative to the control cells. Data are shown as mean ± SD from three independent experiments. In the present study, stringent auto-validation criteria that always provide high confidence of searched results were carried out for protein identification. Overall, more than 300 proteins were identified from each of the three independent experiments conducted. Among these proteins obtained by auto-validation searching criteria, proteins with significant changes (> 1.2- or < 0.8-fold) were classified into distinct categories according to their cellular functions. These included 5 proteins involved in redox regulation of the cell (Fig. 4), 7 proteins participated in cytoskeleton formation and calcium-related signaling pathway (Table 1), 10 metabolic enzymes (Table 2), and 8 other growth-related proteins (Table 3). TABLE 1 List of Differentially Expressed Calcium-Binding Proteins and Cytoskeletal Proteins in HepG2 Cells Incubated With Oxidized SWCNTs and GOa Accession no.  Name  Score  No. of peptides  % AA coverage  (115/114) ± SD  (116/114) ± SD  P07737  Profilin-1  69  5  42  1.37 ± 0.09  1.09 ± 0.08  P21333  Filamin-A  197  16  8  1.35 ± 0.06  1.04 ± 0.11  P31949  Protein S100-A11  24  2  17  1.20 ± 0.18  1.03 ± 0.03  P06703  Protein S100-A6  35  3  22  0.61 ± 0.14  0.87 ± 0.08  P04083  Annexin A1  51  4  17  1.43 ± 0.08  1.30 ± 0.12  P08758  Annexin A5  87  7  23  1.21 ± 0.15  1.13 ± 0.08  P62158  Calmodulin  27  2  11  0.80 ± 0.12  0.87 ± 0.07  Accession no.  Name  Score  No. of peptides  % AA coverage  (115/114) ± SD  (116/114) ± SD  P07737  Profilin-1  69  5  42  1.37 ± 0.09  1.09 ± 0.08  P21333  Filamin-A  197  16  8  1.35 ± 0.06  1.04 ± 0.11  P31949  Protein S100-A11  24  2  17  1.20 ± 0.18  1.03 ± 0.03  P06703  Protein S100-A6  35  3  22  0.61 ± 0.14  0.87 ± 0.08  P04083  Annexin A1  51  4  17  1.43 ± 0.08  1.30 ± 0.12  P08758  Annexin A5  87  7  23  1.21 ± 0.15  1.13 ± 0.08  P62158  Calmodulin  27  2  11  0.80 ± 0.12  0.87 ± 0.07  a 115/114 is the ratio of different protein expression level in the cells treated with oxidized SWCNTs relative to the control cells; 116/114 is the ratio of different protein expression level in the cells treated with GO relative to the control cells. View Large TABLE 2 List of Differentially Expressed Metabolic Enzymes in HepG2 Cells Incubated With Oxidized SWCNTs and GOa Accession no.  Name  Score  No. of peptides  % AA coverage  (115/114) ± SD  (116/114) ± SD  P00558  Phosphoglycerate kinase 1  193  16  37  0.80 ± 0.12  1.13 ± 0.16  P07205  Phosphoglycerate kinase 2  67  6  13  0.76 ± 0.06  0.93 ± 0.04  P15531  Nucleoside diphosphate kinase A  67  5  31  1.26 ± 0.08  0.98 ± 0.10  P00338  L-lactate dehydrogenase A  164  13  39  1.35 ± 0.14  0.94 ± 0.12  P04406  Glyceraldehyde-3-phosphate dehydrogenase  111  9  31  1.52 ± 0.08  0.97 ± 0.06  P21266  Glutathione S-transferase Mu 3  86  6  31  0.54 ± 0.19  0.93 ± 0.12  P23284  Peptidyl-prolyl cis-trans isomerase B  57  5  18  1.46 ± 0.06  0.97 ± 0.15  P23921  Ribonucleoside diphosphate reductase  127  9  14  1.29 ± 0.07  0.92 ± 0.04  Q15084  Protein disulfide-isomerase A6  64  5  15  1.66 ± 0.12  1.15 ± 0.04  P05187  Alkaline phosphatase  27  2  3  1.83 ± 0.14  1.34 ± 0.02  Accession no.  Name  Score  No. of peptides  % AA coverage  (115/114) ± SD  (116/114) ± SD  P00558  Phosphoglycerate kinase 1  193  16  37  0.80 ± 0.12  1.13 ± 0.16  P07205  Phosphoglycerate kinase 2  67  6  13  0.76 ± 0.06  0.93 ± 0.04  P15531  Nucleoside diphosphate kinase A  67  5  31  1.26 ± 0.08  0.98 ± 0.10  P00338  L-lactate dehydrogenase A  164  13  39  1.35 ± 0.14  0.94 ± 0.12  P04406  Glyceraldehyde-3-phosphate dehydrogenase  111  9  31  1.52 ± 0.08  0.97 ± 0.06  P21266  Glutathione S-transferase Mu 3  86  6  31  0.54 ± 0.19  0.93 ± 0.12  P23284  Peptidyl-prolyl cis-trans isomerase B  57  5  18  1.46 ± 0.06  0.97 ± 0.15  P23921  Ribonucleoside diphosphate reductase  127  9  14  1.29 ± 0.07  0.92 ± 0.04  Q15084  Protein disulfide-isomerase A6  64  5  15  1.66 ± 0.12  1.15 ± 0.04  P05187  Alkaline phosphatase  27  2  3  1.83 ± 0.14  1.34 ± 0.02  a 115/114 is the ratio of different protein expression level in the cells treated with oxidized SWCNTs relative to the control cells; 116/114 is the ratio of different protein expression level in the cells treated with GO relative to the control cells. View Large TABLE 3 List of Differentially Expressed Growth-Related Proteins in HepG2 Cells Incubated With Oxidized SWCNTs and GOa Accession no.  Name  Score  No. of peptides  % AA coverage  (115/114) ± SD  (116/114) ± SD  P22626  Nuclear ribonucleoproteins A2/B1  132  12  37  1.25 ± 0.14  1.04 ± 0.06  P62820  Ras-related protein Rab-1A  26  2  13  0.81 ± 0.05  0.97 ± 0.09  P49411  Elongation factor Tu  65  5  13  1.42 ± 0.14  1.05 ± 0.06  P11021  78-kDa glucose-regulated protein  173  13  24  1.20 ± 0.04  1.09 ± 0.02  Q9UQ80  Proliferation-associated protein 2G4  27  2  5  1.27 ± 0.10  1.17 ± 0.02  P63208  S-phase kinase-associated protein 1  24  2  13  0.47 ± 0.22  0.76 ± 0.17  P10809  60-kDa heat shock protein  240  18  33  1.34 ± 0.16  1.01 ± 0.02  P0C0S8  Histone H2A type 1  46  3  23  1.50 ± 0.16  1.04 ± 0.12  Accession no.  Name  Score  No. of peptides  % AA coverage  (115/114) ± SD  (116/114) ± SD  P22626  Nuclear ribonucleoproteins A2/B1  132  12  37  1.25 ± 0.14  1.04 ± 0.06  P62820  Ras-related protein Rab-1A  26  2  13  0.81 ± 0.05  0.97 ± 0.09  P49411  Elongation factor Tu  65  5  13  1.42 ± 0.14  1.05 ± 0.06  P11021  78-kDa glucose-regulated protein  173  13  24  1.20 ± 0.04  1.09 ± 0.02  Q9UQ80  Proliferation-associated protein 2G4  27  2  5  1.27 ± 0.10  1.17 ± 0.02  P63208  S-phase kinase-associated protein 1  24  2  13  0.47 ± 0.22  0.76 ± 0.17  P10809  60-kDa heat shock protein  240  18  33  1.34 ± 0.16  1.01 ± 0.02  P0C0S8  Histone H2A type 1  46  3  23  1.50 ± 0.16  1.04 ± 0.12  a 115/114 is the ratio of different protein expression level in the cells treated with oxidized SWCNTs relative to the control cells; 116/114 is the ratio of different protein expression level in the cells treated with GO relative to the control cells. View Large FIG. 5. View largeDownload slide Relative cell proliferation rate and intracellular ROS level for the HepG2 cells treated with oxidized SWCNTs and GO. Cell proliferation was measure after 48 h treatment of oxidized SWCNTs and GO using MTT assay. Intracellular ROS level was recorded after 12 h treatment of oxidized SWCNTs and GO. Data are shown as mean ± SD from three independent experiments and p values are less than 0.05 over the control. FIG. 5. View largeDownload slide Relative cell proliferation rate and intracellular ROS level for the HepG2 cells treated with oxidized SWCNTs and GO. Cell proliferation was measure after 48 h treatment of oxidized SWCNTs and GO using MTT assay. Intracellular ROS level was recorded after 12 h treatment of oxidized SWCNTs and GO. Data are shown as mean ± SD from three independent experiments and p values are less than 0.05 over the control. Differentially Expressed Proteins Involved in Redox Regulation of the Cell As shown in Figure 4, we successfully identified five proteins involved in redox regulation of the cell. These included peroxiredoxin (Prx) family proteins and thioredoxin. Prx family proteins are thioredoxin peroxidases involved directly in eliminating hydrogen peroxide and neutralizing other ROS (Nordberg and Arnér, 2001; Sue et al., 2005). Specifically, Prx-1, Prx-2, and Prx-6 were significantly upregulated by 1.21- to 1.38-fold for the cells treated with oxidized SWCNTs when compared with the control cells (Fig. 4). Another protein involved in redox regulation of the cell, thioredoxin, was also found to be elevated by 1.20-fold. The upregulation of these Prx proteins and thioredoxin for the cells treated with oxidized SWCNTs suggested that the cells might be under oxidative stress after exposure to oxidized SWCNTs. In comparison, the relative abundance of these Prx proteins and thioredoxin for the batch treated with GO did not show any significant change over the control cells (Fig. 4). Interestingly, Prx-5 was found to be elevated for GO-treated cells, but it was downregulated for the oxidized SWCNT-treated cells. Because previous investigations have demonstrated that Prx-5 is a thioredoxin peroxidase that inhibits p53-induced apoptosis (Kropotov et al., 2006; Zhou et al., 2000), the downregulation of Prx-5 might trigger the activation of p53-mediated DNA damage checkpoint signals and lead to apoptosis when the DNA damage appeared to be irreparable. Differentially Expressed Cytoskeletal Proteins and Signaling Proteins Among the proteins listed in Table 1, two cytoskeleton related proteins, namely profilin and filamin, were found to be upregulated by approximately 35% for the batch of cells treated with oxidized SWCNTs. Specifically, profilin is a ubiquitous actin monomer-binding protein that involved in regulating actin polymerization in response to extracellular signals (Mahoney et al., 1997), and actin-binding protein filamin functions as cross-linkers between plasma membranes and actin-based cytoskeletons. The upregulation of these cytoskeleton proteins suggested that the exposure of oxidized SWCNTs might alter the intracellular microfilament network. In contrast, no obvious effect on these cytoskeletal proteins for GO-treated cells was observed, which indicated the weak interactions of GO with the intracellular cytoskeleton system. As listed in Table 1, several calcium-binding proteins involved in signaling pathways were found to be differentially expressed. S100 proteins are involved in regulation of protein phosphorylation, transcription factors, calcium homeostasis, the dynamics of cytoskeleton constituents, cell growth, and the inflammatory response (Donato, 2003; Zimmer et al., 1995). Interestingly, S100-A11 that deters the cell growth and S100-A6 that promotes cell growth were found to be 1.20- and 0.61-fold for the batch treated with oxidized SWCNTs. As the reciprocal regulation of protein S100-A6 and S100-A11 determines the metabolic rate and cell proliferation rate, our finding suggested that the cell growth might be severely deterred after exposure to oxidized SWCNTs. Furthermore, calmodulin, a ubiquitous intracellular Ca2+ receptor that regulates various cellular functions (Chin and Means, 2000), was also found to be downregulated by 20% for oxidized SWCNT-treated cells. Because various calmodulin-dependent protein kinases need to be activated in proliferating cells, the downregulation of calmodulin further supported the exposure of oxidized SWCNTs might deter the cell proliferation rate. In contrast, these S100 proteins showed only moderate variations for the GO-treated cells as protein S100-A11 and S100-A6 appeared to be 1.03- and 0.87-fold of the control cells. Moreover, we found the expression levels of annexins, another family of calcium-binding proteins, were elevated for GO- and SWCNT-treated cells. As annexins have been involved in various cellular activities such as signal transduction, endocytosis, signal organization of the extracellular matrix, and resistance to ROS (Benz and Hofmann, 1997; Laohavisit and Davies, 2011), their exact role in modulating the cellular functions in response to these nanomaterials remains to be further studied. Differentially Expressed Metabolic Enzymes A close analysis of the metabolic enzymes listed in Table 2 revealed a distinct pattern of changes between oxidized SWCNTs and GO-treated cells, which suggested different modes of action for these nanomaterials. Specifically, two key metabolic enzymes involved in glycolysis were found to be significantly altered for the batch of cells treated with oxidized SWCNTs. These included glyceraldehydes 3-phosphate dehydrogenase that converts glyceraldehyde 3-phosphate to 1,3-bisphosphoglycerate, and phosphoglycerate kinase that transfers a phosphate group from 1,3-bisphosphoglycerate to ADP, forming ATP and 3-phosphoglycerate. In contrast, these metabolic enzymes for the batch of cells treated with GO appeared to be at a similar level as in the control cells, which indicated that the cellular metabolism was not perturbed by the GO exposure. Moreover, we found glutathione S-transferase Mu (GSTMu), involved in detoxification of endogenous compounds through conjugating of reduced glutathione (Patskovsky et al., 1999), was significantly downregulated by 50% for the cells exposed to oxidized SWCNTs. Peptidyl-prolyl cis-trans isomerase that functions as a protein-folding chaperone was found to be 1.46-fold of the control cells for the oxidized SWCNT-treated cells, but not for the GO-treated cells. Another enzyme that acts to catalyze protein folding, protein disulfide isomerase A6, was also found to be upregulated by 1.66-fold for the cells exposed to oxidized SWCNTs only. Furthermore, the expression levels of various other metabolic enzymes listed in Table 2 were found to be significantly altered for the batch of cells treated with oxidized SWCNTs. Differentially Expressed Other Growth-Related Proteins To further strengthen the hypothesis that GO might be more biocompatible in comparison to oxidized SWCNTs, we examined other growth-related proteins as listed in Table 3. We found 60-kDa heat shock protein and 78-kDa glucose-regulated protein were significantly upregulated with iTRAQ ratios by 20–34% for the cells treated with oxidized SWCNTs. In contrast, these proteins remained at the same level as the control cells for the GO-treated cells. A closer analysis of the proteins listed in Table 3 further revealed that the expression levels of ribonucleoproteins, histone, and elongation factor also appeared to be much higher for the cells exposure to oxidized SWCNTs over GO. Moreover, proliferation-associated protein 2G4, an RNA-binding protein involved in growth regulation, was found to be significantly upregulated by 1.27-fold for the oxidized SWCNT-treated cells. Another cell cycle control protein, S-phase kinase-associated protein 1 that controls the progression from G1 phase to S phase of the mitotic cell cycle, was significantly downregulated by 54% for the batch of cells treated with oxidized SWCNTs. The elevated level of proliferation-associated protein 2G4 and the downregulation of S-phase kinase-associated protein 1 suggested that the growth rate and/or the cell cycle might be disturbed after exposure to oxidized SWCNTs. In contrast, all these proteins mentioned in this section for the GO-treated cells showed only moderate changes or remained at similar levels as the control cells, which suggested that GO might cause less cytotoxicity to the cells. Oxidized SWCNTs Reduced Cell Proliferation Rate and Induced ROS Generation To determine whether proliferation rate of the human HepG2 cells was altered by the exposure of these nanomaterials, MTT assay that evaluates of cell proliferation rate based on the ability of the mitochondrial dehydrogenase to convert the yellow dye (MTT) to a purple-colored formazan was carried out. The potential toxic effect on cell proliferation can be calculated by comparing the amount of purple formazan produced by the cells treated with nanomaterials over the amount of formazan formed by the untreated cells. After 48 h of exposure to either oxidized SWCNTs or GO, we observed that both cells treated with nanomaterials showed a decrease in metabolic activities when compared with the control batch (Fig. 5). Specifically, we found the cells treated with oxidized SWCNTs lost approximately 18% of metabolic activity and the cells exposed to GO only lost 6% based on the MTT assay result. To determine the effect of oxidized SWCNTs and GO on ROS generation in human HepG2 cells, the fluorescent probe DFDA was used to detect cell-derived oxidants. After exposure to oxidized SWCNTs and GO for 12 h, DFDA was added to the cell culture and the fluorescein formed by the cells was measured after 1 h incubation with DFDA in the dark. As can be seen in Figure 5, the fluorescence intensity of fluorescein produced for the cells exposed to oxidized SWCNTs was higher in comparison to the control batch, which was approximately 1.14-fold over the control cells. However, the fluorescence intensity for the cells treated with GO showed only an 8% increase over the control cells. These data indicated that oxidized SWCNTs induced higher level of ROS in human HepG2 cells in comparison to GO, which might explain more severe oxidative stress in these cells over GO as indicated from the protein profile. Oxidized SWCNTs Induced Apoptosis and Led to Growth Arrest in Human HepG2 Cells To investigate whether human HepG2 cells undergo apoptosis after the treatment of oxidized SWCNTs and GO, cells treated with either of the nanomaterials were examined using annexin V-FITC/PI apoptosis detection assay. Because the appearance of phosphatidyl serine (PS) residues on the surface of the cell is a signal of apoptosis before nuclear breakdown, DNA fragmentation, and the appearance of apoptosis-associated molecules, annexin V-FITC that have a high affinity for binding to PS has been widely used to detect apoptotic cells in the early stage. As can been seen in Figure 6A, we observed a significant increase in the proportion of cells undergoing apoptosis for the batch treated with oxidized SWCNTs. Specifically, the subpopulation of cells staining with annexin V-FITC for cells exposure to oxidized SWCNTs was 26.3% of total cells in comparison to 18.6% of total cells for the control batch. In comparison, there was no significant increase in the proportion of apoptotic cells for the GO-treated batch (Fig. 6A). FIG. 6. View largeDownload slide Assessment of apoptosis in human HepG2 cells after exposure to oxidized SWCNTs and GO (A). Cell cycle distribution analysis of the HepG2 cells after exposure to oxidized SWCNTs and GO (B). Apoptosis data are representative of three independent experiments. The quantitative data for cell distribution are shown as mean ± SD from three independent experiments. FIG. 6. View largeDownload slide Assessment of apoptosis in human HepG2 cells after exposure to oxidized SWCNTs and GO (A). Cell cycle distribution analysis of the HepG2 cells after exposure to oxidized SWCNTs and GO (B). Apoptosis data are representative of three independent experiments. The quantitative data for cell distribution are shown as mean ± SD from three independent experiments. As shown in Figure 6B, cell cycle analysis showed that treatment of human HepG2 cells with oxidized SWCNTs led to a significant increase in the proportion of cells in the G2/M phase. Specifically, the proportion of cells at G2/M phase was 40.3% over 31.5% of the total cells for the control batch. In contrast, the treatment of human HepG2 cells with GO resulted in a minor decrease in the percentage of G2/M phase cells over the control cells. DISCUSSION Because of their distinctive physical and chemical properties, graphitic nanomaterials have attracted tremendous interest in the fields of biotechnology. Oxidized SWCNTs and GO are functionalized materials with improved solubility, which are promising candidate for biomedical applications, such as biosensors and drug carriers (Liu et al., 2010; Sun et al., 2008; Veetil and Ye, 2009). In such scenario, these nanomaterials are in constant interaction with living cells and tissues, which raises several concerns about their biocompatibility. With the growing potential applications in biomedical areas, it is urgently needed to evaluate the potential risks that might be introduced by these nanomaterials and have a fundamental understanding of the interaction between these nanomaterials and living systems. In the present study, our group applied the iTRAQ-coupled 2D LC-MS/MS proteomics approach to study the cellular functions of human hepatoma HepG2 cells in response to oxidized SWCNTs and GO. We successfully identified 30 differentially expressed proteins in cells either treated with GO or oxidized SWCNTs, which provided useful information to evaluate the potential cytotoxicity of these nanomaterials and elucidate the underlying cytotoxicity mechanisms. Because previous investigations indicated that CNTs could be taken by mammalian cells (Cheng et al., 2009; Porter et al., 2009; Ye et al., 2011), the protein profile obtained by iTRAQ-based proteomics approach suggested that these oxidized SWCNTs retained inside the cells might interfere with metabolic routes, protein synthesis, mRNA processing, cytoskeleton system, and signaling pathways. In contrast, only moderate variation of the protein profile was observed for the cells exposed to GO, which suggested less cytotoxicity of GO-based nanomaterials. By applying iTRAQ-coupled 2D LC-MS/MS analysis, we successfully identified changes in the levels of key proteins involved in the redox regulation of the cell. Specifically, Prx family protein and thioredoxin were found to be significantly upregulated in the cells treated with oxidized SWCNTs, which suggested there might be an elevated level of intracellular ROS after treatment with oxidized SWCNTs. Further measurement of the intracellular ROS level confirmed that SWCNTs induced higher level of ROS in comparison to the control cells (Fig. 5), which indicated that cells after exposure to oxidized SWCNTs were undergoing oxidative stress. However, Prx-5, a thioredoxin peroxidase that inhibits p53-induced apoptosis (Kropotov et al., 2006; Zhou et al., 2008), was found to be significantly downregulated by 21% for the cells exposed to oxidized SWCNTs. Its downregulation suggested that oxidized SWCNTs might activate p53-mediated DNA damage checkpoint signals and lead to apoptosis. Because p53 can also induce growth arrest by holding the cell cycle at the G1/S and G2/M regulation point on DNA damage recognition (Agarwal et al., 1995), it might explain the significant increase in the proportion of G2/M phase cells for oxidized SWCNT-treated cells as observed from the cell cycle analysis (Fig. 6B). These findings were consistent with a previous study showing that growth arrest at G2/M phase was observed for the oxidized SWCNT-treated human primary monocytes (Ye et al., 2011). Moreover, we found GSTMu that modulates the stress-activated signals by suppressing apoptosis signal-regulating kinase 1 (ASK1) (Cho et al., 2001) was significantly downregulated by 50% after exposure to oxidized SWCNTs. Its downregulation might induce the activation of ASK1, and therefore resulted in the activation of downstream signaling cascades, such as the stress-associated protein kinases and p38 pathway, and finally led to stress-induced apoptosis. From the apoptosis analysis result (Fig. 6A), we did find a significant increase in the proportion of cells undergoing apoptosis after the treatment of oxidized SWCNTs; however, further experiments are needed to confirm the hypothesis that observed apoptosis was mediated through the ASK1 stress-induced pathway. Moreover, as previous investigation has demonstrated that the stimulation of p38 activates the nuclear factor kappa B (NF-κB) (Craig et al., 2000), our findings suggested that the activation of transcription factor NF-κB by oxidized SWCNTs exposure reported earlier might be through p38 signaling pathway and the collaborative action of these signals might also explain the interleukin-6/8–mediated inflammatory response in human primary monocytes (Ye et al., 2011). Furthermore, S100-A6 and calmodulin that promote cell growth were found to be 0.61- and 0.80-fold of the control cells for the batch treated with oxidized SWCNTs. S100-A11 protein that deters the cell growth was found to be elevated by 20% in comparison to the control cells. Because the reciprocal regulation of these growth rate control proteins decides the cell proliferation rate, our findings suggested that the growth rate of the cells might be reduced after exposure to oxidized SWCNTs. Further evidence on reduced proliferation rate could also be obtained from the downregulation of the metabolic enzyme, such as phosphoglycerate kinase involved in the ATP generating step, MTT assay data, and cell cycle analysis results. Furthermore, 60-kDa heat shock protein that functions as intracellular chaperones and assists protein folding (Schlesinger, 1990) was found to be significantly upregulated by 34% for the cells treated with oxidized SWCNTs; 78-kDa glucose-regulated protein, one of the endoplasmic reticulum (ER) chaperones and a key marker of ER stress, was found to be elevated by 20% after the treatment of oxidized SWCNTs. Peptidyl-prolyl cis-trans isomerase that functions as protein-folding chaperone was found to be 1.46-fold of the control and another enzyme that acts to catalyze protein-folding, protein disulfide isomerase A6, was also found to be upregulated by 66% for the cells exposed to oxidized SWCNTs. The significant upregulation of these chaperone proteins involved in assisting protein folding might be explained by elevated protein demand as various proteins were found to be significantly upregulated after the treatment of oxidized SWCNTs and/or oxidized SWCNTs might perturb the protein-folding process. Moreover, the cell cycle regulation proteins, such as proliferation-associated protein 2G4 and S-phase kinase-associated protein 1 that controls the progression from G1 phase to S phase of the cell cycle, were found to be 1.27- and 0.47-fold of the control cells, respectively. These differentially expressed growth regulation proteins provided further evidence that oxidized SWCNTs could perturb the growth rate and cell cycle as discussed above. A closer analysis of the proteins listed in Table 3 revealed that the expression levels of other growth related proteins, such as ribonucleoproteins, histone, and elongation factor, also appeared to be much higher for the cells exposure to oxidized SWCNTs, which indicated that oxidized SWCNTs had profound effects on the intracellular systems. Based on the information obtained from protein profile, GO showed moderate effects on the cellular functions in comparison with oxidized SWCNTs, which confirmed low cytotoxicity of GO as previously suggested (Hu et al., 2010; Ryoo et al., 2010). From the protein profile, we found S100-A11 and S100-A6 and calmodulin that promote cell growth were downregulated by approximately 13% over the control cells. In addition, other cell growth rate regulation proteins, such as proliferation-associated protein 2G4 and S-phase kinase-associated protein 1, were also found to be altered by the exposure of GO. These findings suggested that growth rate and cell cycle might still be perturbed after the treatment of GO. Such a hypothesis was further confirmed by functional assays as we did found a minor reduction in proliferation rate after the cells exposed to GO based on the MTT assay result (Fig. 5) and cell cycle was also slightly perturbed (Fig. 6B). Moreover, elevated intracellular ROS level was also observed for GO-treated cells. However, cellular responses between oxidized SWCNTs and GO-treated cells were significantly different as we only found Prx-5 in the Prx family proteins was upregulated in the GO-treated cells. The distinct pattern of these Prx proteins might lead to a distinct fate of the cells as these proteins were involved in mediating signal transduction and modulating various cellular activities in mammalian cells. Because it is nearly impossible to introduce the same degree of oxidization for these nanomaterials, it might cause some differences in the interaction with the intracellular systems. However, these subtle differences during the functionalization could hardly explain the distinct pattern of protein profiles obtained here. Previously, Cheng et al. have reported that MWCNTs entered the cell both actively and passively frequently inserting through the plasma membrane into the cytoplasm and the nucleus (Cheng et al., 2009). Considering the similarity between MWCNTs and oxidized SWCNTs regarding their physical properties, we expected that oxidized SWCNTs might also cause incomplete phagocytosis or mechanically pierce through the plasma membrane and result in oxidative stress and cell death through apoptosis in a similar manner like MWCNTs. As GO nanosheets have also been reported to be internalized within mammalian cells by endocytosis and retained inside the endosome of the cytoplasm (Hu et al., 2010), the low cytotoxicity of GO might be explained by the poor puncturing effect and the compartmentation of GO inside the endosome significantly prevented the severe interference with other intracellular systems. In summary, our group has demonstrated the effectiveness of applying iTRAQ-coupled 2D LC-MS/MS proteome analysis to study the cellular functions of HepG2 cells in response to nanomaterials. By the close look at cellular functions at the proteome level, we clearly identified the distinct pattern of cellular responses between GO and oxidized SWCNT-treated cells, which suggested different mode of actions to the living systems regarding these nanomaterials. Further functional assays confirmed that oxidized SWCNTs triggered elevated level of ROS, reduced cellular metabolic activity, perturbed cell cycle, and led to a significantly higher proportion of apoptotic cells. We envision that the systematic characterization through comparative proteomics technique coupled with functional assays would provide us a more detailed understanding of the interactions between living systems and these nanomaterials. FUNDING School of Chemical and Biomedical Engineering, Nanyang Technological University, Singapore (J.Y. and H.G.). References Agarwal ML Agarwal A Taylor WR Stark GR p53 Controls both the G2/M and the G1 cell cycle checkpoints and mediates reversible growth arrest in human fibroblasts Proc. Natl. Acad. Sci. U.S.A.  1995 92 8493 8497 Google Scholar CrossRef Search ADS PubMed  Benz J Hofmann A Annexins: From structure to function Biol. Chem.  1997 378 177 183 Google Scholar PubMed  Chen JL Yan XP A dehydration and stabilizer-free approach to production of stable water dispersions of graphene nanosheets J. Mater. Chem.  2010 20 4328 4332 Google Scholar CrossRef Search ADS   Cheng C Müller KH Koziol KKK Skepper JN Midgley PA Welland ME Porter AE Toxicity and imaging of multi-walled carbon nanotubes in human macrophage cells Biomaterials  2009 30 4152 4160 Google Scholar CrossRef Search ADS PubMed  Chin D Means AR Calmodulin: A prototypical calcium sensor Trends Cell Biol.  2000 10 322 328 Google Scholar CrossRef Search ADS PubMed  Cho SG Lee YH Park HS Ryoo K Kang KW Park J Eom SJ Kim MJ Chang TS Choi SY et al.   Glutathione S-transferase mu modulates the stress-activated signals by suppressing apoptosis signal-regulating kinase 1 J. Biol. Chem.  2001 276 12749 12755 Google Scholar CrossRef Search ADS PubMed  Choi BG Park H Park TJ Yang MH Kim JS Jang SY Heo NS Lee SY Kong J Hong WH Solution chemistry of self-assembled graphene nanohybrids for high-performance flexible biosensors ACS Nano  2010 4 2910 2918 Google Scholar CrossRef Search ADS PubMed  Chou CC Hsiao HY Hong QS Chen CH Peng YW Chen HW Yang PC Single-walled carbon nanotubes can induce pulmonary injury in mouse model Nano Lett.  2008 8 437 445 Google Scholar CrossRef Search ADS PubMed  Craig R Larkin A Mingo AM Thuerauf DJ Andrews C McDonough PM Glembotski CC p38 MAPK and NF-κB collaborate to induce interleukin-6 gene expression and release: Evidence for a cytoprotective autocrine signaling pathway in a cardiac myocyte model system J. Biol. Chem.  2000 275 23814 23824 Google Scholar CrossRef Search ADS PubMed  Ding L Stilwell J Zhang T Elboudwarej O Jiang H Selegue JP Cooke PA Gray JW Chen FF Molecular characterization of the cytotoxic mechanism of multiwall carbon nanotubes and nano-onions on human skin fibroblast Nano Lett.  2005 5 2448 2464 Google Scholar CrossRef Search ADS PubMed  Donato R Intracellular and extracellular roles of S100 proteins Microsc. Res. Tech.  2003 60 540 551 Google Scholar CrossRef Search ADS PubMed  Hempel SL Buettner GR O'Malley YQ Wessels DA Flaherty DM Dihydrofluorescein diacetate is superior for detecting intracellular oxidants: Comparison with 2′,7′-dichlorodihydrofluorescein diacetate, 5(and 6)-carboxy-2',7'-dichlorodihydrofluorescein diacetate, and dihydrorhodamine 123 Free Radic. Biol. Med.  1999 27 146 159 Google Scholar CrossRef Search ADS PubMed  Hu W Peng C Luo W Lv M Li X Li D Huang Q Fan C Graphene-based antibacterial paper ACS Nano  2010 4 4317 4323 Google Scholar CrossRef Search ADS PubMed  Kovtyukhova NI Layer-by-layer assembly of ultrathin composite films from micron-sized graphite oxide sheets and polycations Chem. Mater.  1999 11 771 778 Google Scholar CrossRef Search ADS   Kropotov A Serikov V Suh J Smirnova A Bashkirov V Zhivotovsky B Tomilin N Constitutive expression of the human peroxiredoxin V gene contributes to protection of the genome from oxidative DNA lesions and to suppression of transcription of noncoding DNA FEBS J.  2006 273 2607 2617 Google Scholar CrossRef Search ADS PubMed  Laohavisit A Davies JM Annexins New Phytol.  2011 189 40 53 Google Scholar CrossRef Search ADS PubMed  Liu Y Yu D Zeng C Miao Z Dai L Biocompatible graphene oxide-based glucose biosensors Langmuir  2010 26 6158 6160 Google Scholar CrossRef Search ADS PubMed  Liu Z Chen K Davis C Sherlock S Cao Q Chen X Dai H Drug delivery with carbon nanotubes for in vivo cancer treatment Cancer Res.  2008 68 6652 6660 Google Scholar CrossRef Search ADS PubMed  Magrez A Kasas S Salicio V Pasquier N Seo JW Celio M Catsicas S Schwaller B Forró L Cellular toxicity of carbon-based nanomaterials Nano Lett.  2006 6 1121 1125 Google Scholar CrossRef Search ADS PubMed  Mahoney NM Janmey PA Almo SC Structure of the profilin-poly-L-proline complex involved in morphogenesis and cytoskeletal regulation Nat. Struct. Biol.  1997 4 953 960 Google Scholar CrossRef Search ADS PubMed  Manna SK Sarkar S Barr J Wise K Barrera EV Jejelowo O Rice-Ficht AC Ramesh GT Single-walled carbon nanotube induces oxidative stress and activates nuclear transcription factor-κB in human keratinocytes Nano Lett.  2005 5 1676 1684 Google Scholar CrossRef Search ADS PubMed  Mu Q Li Z Li X Mishra SR Zhang B Si Z Yang L Jiang W Yan B Characterization of protein clusters of diverse magnetic nanoparticles and their dynamic interactions with human cells J. Phys. Chem. C  2009 113 5390 5395 Google Scholar CrossRef Search ADS   Nordberg J Arnér ESJ Reactive oxygen species, antioxidants, and the mammalian thioredoxin system Free Radic. Biol. Med.  2001 31 1287 1312 Google Scholar CrossRef Search ADS PubMed  Patskovsky YV Patskovska LN Listowsky I An asparagine-phenylalanine substitution accounts for catalytic differences between hGSTM3-3 and other human class Mu glutathione S-transferases Biochemistry  1999 38 16187 16194 Google Scholar CrossRef Search ADS PubMed  Porter AE Gass M Bendall JS Muller K Goode A Skepper JN Midgley PA Welland M Uptake of noncytotoxic acid-treated single-walled carbon nanotubes into the cytoplasm of human macrophage cells ACS Nano  2009 3 1485 1492 Google Scholar CrossRef Search ADS PubMed  Pulskamp K Diabaté S Krug HF Carbon nanotubes show no sign of acute toxicity but induce intracellular reactive oxygen species in dependence on contaminants Toxicol. Lett.  2007 168 58 74 Google Scholar CrossRef Search ADS PubMed  Raja PMV Connolley J Ganesan GP Ci L Ajayan PM Nalamasu O Thompson DM Impact of carbon nanotube exposure, dosage and aggregation on smooth muscle cells Toxicol. Lett.  2007 169 51 63 Google Scholar CrossRef Search ADS PubMed  Robinson JT Perkins FK Snow ES Wei Z Sheehan PE Reduced graphene oxide molecular sensors Nano Lett.  2008 8 3137 3140 Google Scholar CrossRef Search ADS PubMed  Ross PL Huang YN Marchese JN Williamson B Parker K Hattan S Khainovski N Pillai S Dey S Daniels S et al.   Multiplexed protein quantitation in Saccharomyces cerevisiae using amine-reactive isobaric tagging reagents Mol. Cell. Proteomics  2004 3 1154 1169 Google Scholar CrossRef Search ADS PubMed  Ryoo SR Kim YK Kim MH Min DH Behaviors of NIH-3T3 fibroblasts on graphene/carbon nanotubes: Proliferation, focal adhesion, and gene transfection studies ACS Nano  2010 4 6587 6598 Google Scholar CrossRef Search ADS PubMed  Schlesinger MJ Heat shock proteins J. Biol. Chem.  1990 265 12111 12114 Google Scholar PubMed  Shvedova AA Castranova V Kisin ER Schwegler-Berry D Murray AR Gandelsman VZ Maynard A Baron P Exposure to carbon nanotube material: Assessment of nanotube cytotoxicity using human keratinocyte cells J. Toxicol. Environ. Health A  2003 66 1909 1926 Google Scholar CrossRef Search ADS PubMed  Sue GR Ho ZC Kim K Peroxiredoxins: A historical overview and speculative preview of novel mechanisms and emerging concepts in cell signaling Free Radic. Biol. Med.  2005 38 1543 1552 Google Scholar CrossRef Search ADS PubMed  Sun XM Liu Z Welsher K Robinson JT Goodwin A Zaric S Dai HJ Nano-graphene oxide for cellular imaging and drug delivery Nano Res.  2008 1 203 212 Google Scholar CrossRef Search ADS PubMed  Veetil JV Ye K Tailored carbon nanotubes for tissue engineering applications Biotechnol. Prog.  2009 25 709 721 Google Scholar CrossRef Search ADS PubMed  Wu ZS Ren W Gao L Liu B Jiang C Cheng HM Synthesis of high-quality graphene with a pre-determined number of layers Carbon  2009 47 493 499 Google Scholar CrossRef Search ADS   Ye S Zhang H Wang Y Jiao F Lin C Zhang Q Carboxylated single-walled carbon nanotubes induce an inflammatory response in human primary monocytes through oxidative stress and NF-κB activation J. Nanoparticle Res.  2011 13 4239 4252 Google Scholar CrossRef Search ADS   Zhang D Yi C Zhang J Chen Y Yao X Yang M The effects of carbon nanotubes on the proliferation and differentiation of primary osteoblasts Nanotechnology  2007 18 475102 475111 Google Scholar CrossRef Search ADS   Zhang T Stilwell JL Gerion D Ding L Elboudwarej O Cooke PA Gray JW Alivisatos AP Chen FF Cellular effect of high doses of silica-coated quantum dot profiled with high throughput gene expression analysis and high content cellomics measurements Nano Lett.  2006 6 800 808 Google Scholar CrossRef Search ADS PubMed  Zhou H Mu Q Gao N Liu A Xing Y Gao S Zhang Q Qu G Chen Y Liu G et al.   A nano-combinatorial library strategy for the discovery of nanotubes with reduced protein-binding, cytotoxicity, and immune response Nano Lett.  2008 8 859 865 Google Scholar CrossRef Search ADS PubMed  Zhou Y Kok KH Chun ACS Wong CM Wu HW Lin MCM Fung PCW Kung HF Jin DY Mouse peroxiredoxin V is a thioredoxin peroxidase that inhibits p53-induced apoptosis Biochem. Biophys. Res. Commun.  2000 268 921 927 Google Scholar CrossRef Search ADS PubMed  Zimmer DB Cornwall EH Landar A Song W The S100 protein family: History, function, and expression Brain Res. Bull.  1995 37 417 429 Google Scholar CrossRef Search ADS PubMed  © The Author 2011. Published by Oxford University Press on behalf of the RCGP. All rights reserved. For permissions please e-mail: journals.permissions@oup.com
TI - Cytotoxicity Evaluation of Oxidized Single-Walled Carbon Nanotubes and Graphene Oxide on Human Hepatoma HepG2 cells: An iTRAQ-Coupled 2D LC-MS/MS Proteome Analysis
JO - Toxicological Sciences
DO - 10.1093/toxsci/kfr332
DA - 2011-12-08
UR - https://www.deepdyve.com/lp/oxford-university-press/cytotoxicity-evaluation-of-oxidized-single-walled-carbon-nanotubes-and-eH5gJCehrP
SP - 149
EP - 161
VL - 126
IS - 1
DP - DeepDyve
ER -