Differential Nuclear Proteomes in Response to N-Methyl-N′-nitro-N-nitrosoguanidine Exposure Jing Shen, Huifang Zhu, Xueping Xiang, and Yingnian Yu* Department of Pathology and Pathophysiology, Zhejiang University School of Medicine, Hangzhou 310058, China Received January 5, 2009
As a model carcinogen in studying the mutagenicity and carcinogenicity of N-nitroso alkylating agent, N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) can induce DNA damages and many cellular defensive events, such as DNA repair, G2/M-phase arrest and apoptosis. Noticeably, diverse cellular responses were observed to occur following MNNG treatment in our previous studies, including nontargeted mutations (NTM) at undamaged DNA bases, endoplasmic reticulum stress (ER stress) induction and the activation of several signal transduction pathways. In addition, whole cell proteome analysis also revealed that comprehensive and various changes were triggered by this mutagen. However, lowabundance proteins with key functions, such as nuclear proteins, are always underrepresented in proteomic studies. To reduce the complexity of protein samples and monitor subcellular alterations in response to MNNG exposure, nuclear extracts were fractionated from MNNG-treated cells and compared using two-dimensional fluorescence difference gel electrophoresis (2-D DIGE). Twenty-three differentially expressed protein spots were observed after 0.25 and 1 µM MNNG exposure, and 8 of them were shared between these two treatment groups. When MALDI-TOF MS was used, 17 nuclear proteins affected by MNNG were identified. These proteins were involved in regulation of RNA processing, signal transduction, metabolism, DNA repair, protein biosynthesis and microtubule or cytoskeleton organization. Among them, two nuclear proteins with nucleocytoplasmic shuttling activity, 14-3-3 ζ and hnRNP K, were further demonstrated to undergo different dynamic changes in response to MNNG exposure. These results will aid our understanding of the complicated mechanisms of MNNG-induced cellular responses. Keywords: N-Methyl-N′-nitro-N-nitrosoguanidine • Nuclear proteome • Two-dimensional fluorescence difference gel electrophoresis • Mass spectrometry
Introduction Exposure to environmental carcinogens is considered one of the major factors contributing to human cancer risk. Although the relationship between carcinogen exposure and tumor development is extensively studied,1,2 the complex molecular mechanisms and biological pathways involved in tumorigenesis are still poorly understood. As a model carcinogen in studying the mutagenicity and carcinogenicity of Nnitroso alkylating agent, N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) can trigger many important cellular responses, such as DNA repair, G2/M-phase arrest and apoptosis.3,4 Our previous studies also revealed that diverse cellular responses can be induced by this mutagen. For instance, low concentrations of MNNG (0.2∼2 µM) can cause nontargeted mutation (NTM) at undamaged DNA bases, which was demonstrated as a result of the decreased DNA replication fidelity.5,6 Further analyses showed that the expression level of some DNA polymerases with low fidelity in nucleotide insertion was * To whom correspondence should be addressed. Prof. Yingnian Yu, Department of Pathology and Pathophysiology, Zhejiang University School of Medicine, Hangzhou 310058, China. E-mail:
[email protected]. Fax: +86571-88208209. 10.1021/pr900008n CCC: $40.75
2009 American Chemical Society
changed following MNNG treatment.7,8 We also found that the epithelial growth factor receptor (EGF) clustered at the cellular membrane after MNNG exposure, but did not activate Ras, the downstream mediator in EGFR signaling pathway, as compared to EGF treatment.9,10 Additionally, endoplasmic reticulum stress (ER stress) and several signal transduction pathways such as the c-Jun N-terminal Kinase (JNK) and cAMP-PKA-CREB pathways were also observed to be activated after MNNG treatmnet.11-13 Diverse and complicated cellular responses induced by MNNG encouraged us to utilize high-throughput techniques, especially proteomics which has risen to a position of supreme importance in the post-genomic era. We have shown previously that exposure to different concentrations of MNNG can induce comprehensive and various changes in the protein expression profile with the combined 2-DE and MS proteomic approach.14 These proteins belong to a broad range of different classes and functional pathways, such as transcription regulation, protein biosynthesis, signal transduction, cytoskeleton organization, transport and so forth, reflecting the wide variety of cellular processes in which they take part. However, proteomic analysis may be limited by failure to detect low-abundance proteins. One means of overcoming this problem is to fractionate the Journal of Proteome Research 2009, 8, 2863–2872 2863 Published on Web 04/13/2009
research articles protein mixture prior to 2-DE. Subcellular fractionation, or organelle proteomics, reduces the complexity of eukaryotic proteomes, enabling visualization of lower abundance proteins as part of a specific group of proteins central to the biological problem under investigation.15 Nuclear proteins play a central role in controlling cellular functions, including the activation of signal transduction pathways and regulation of gene expression after carcinogen exposure. However, they are often underrepresented in proteomic studies due to their lower abundance in comparison to cellular ‘housekeeping’ metabolic enzymes and structural proteins. 2-D DIGE enables identification and more accurate quantification of differentially expressed proteins, since gelto-gel variation is avoided by running control and test samples with an internal standard for normalization in the same gel.16,17 DIGE has been used to investigate differentially expressed nuclear proteins in a variety of researches.18-20 In this study, we have quantitatively analyzed the nuclear proteome in human amnion epithelial cells exposed to different concentrations of MNNG using DIGE. As reported here, 23 nuclear proteins were affected after MNNG treatment, and 17 of them were successfully identified by MALDI-MS. Some nuclear proteins with nucleocytoplasmic shuttling activity, such as 14-3-3s and hnRNP K, were found to undergo dynamic changes in response to MNNG exposure. Further comprehensive characterization of these proteins will lead to a better understanding of the mechanisms of cellular responses induced by this mutagen.
Materials and Methods Cell Culture and Treatment. The human amniotic epithelial cell (FL cell) (ATCC CCL-62), with typical morphology and epithelial cell biomarker expression, were cultured in Eagle’s minimal essential medium (MEM) supplemented with 10% newborn bovine serum, 100 units/mL penicillin, 100 µg/mL streptomycin and 200 µg/mL kanamycin in a water-saturated, 5% CO2 atmosphere at 37 °C. When cells achieved logarithmic growth, they were exposed to 0.25 or 1 µM MNNG (Sigma, St. Louis, MO) for 2.5 h. Dimethyl sulfoxide (DMSO) was used as solvent control. After complete removal of the chemicals, cells were recovered in fresh medium for another 12 h before lysis. Protein Extraction and Cy-Dye Labeling. Nuclear extraction was performed with 2-D Sample Prep for Nuclear Proteins Kit (Pierce, Rockford, IL) following the manufacturer’s instructions. After further precipitation and purification using 2-D Cleanup Kit (GE Healthcare BioSciences, Little Chalfont, U.K.), the nuclear proteins were solubilized in lysis buffer containing 7 M urea, 2 M thiourea, 4% CHAPS, 30 mM Tris base, 35 µg/mL PMSF, 0.5 µg/mL leupeptin and 2 µg/mL aprotinin. Protein concentration was determined using 2-D Quant Kit (GE Healthcare). Protein samples (50 µg) were labeled with CyDye DIGE fluors (400 pmol) (GE Healthcare) following the manufacturer’s instructions. CyDye labeling was randomized between the 12 nuclear protein samples (three treatment groups including DMSO, 0.25 and 1 µM MNNG, four biological replicates each) to rule out dye-bias. A pooled internal standard containing equal quantities of all 12 nuclear extracts was labeled with Cy2. Labeling was performed for 30 min on ice in the dark, after which the reactions were quenched with the addition of 10 mM lysine for 10 min on ice in the dark. The quenched 50 µg Cy3and Cy5-labeled samples were then combined and mixed with 50 µg of the quenched Cy2-labeled internal standard, after 2864
Journal of Proteome Research • Vol. 8, No. 6, 2009
Shen et al. which an equal volume of 2× sample buffer (7 M urea, 2 M thiourea, 4% CHAPS, 2% DTT) supplemented with 2% Pharmalyte pH 3-10 was added. 2-D Gel Electrophoresis and Imaging. 2-DE was performed in Ettan IPGphor II IEF and DALT II systems (GE Healthcare) according to the manufacturer’s instructions. The Cy-dye labeled protein fractions were diluted with a rehydration solution (7 M urea, 2 M thiourea, 2% CHAPS, 0.4% DTT, 0.5% IPG buffer, pH 4-7, 0.002% bromophenol blue) to 450 µL and applied to 24 cm Immobiline Drystrips (pH 4-7, GE Healthcare). Rehydration and IEF were carried out on the IPGphor as follows: rehydration was begun with 50 V for 12 h, followed by 200 V for 1 h, 500 V for 1 h, 1000 V for 1 h, 1000-8000 V for 30 min, and 8000 V was applied until the total Vh reached 67.0-70.0 kVh at 20 °C. After IEF separation, the strips were equilibrated for 15 min in an equilibration buffer (100 mM TrisHCl, 6 M Urea, 30% glycerol, 2% SDS and 0.002% bromophenol blue) containing 0.5% DTT, then equilibrated again for another 15 min in the same buffer, except that DTT was replaced with 4.5% iodoacetamide. Second-dimension SDS electrophoresis was carried out on 12.5% slab gels using the ETTAN DALT II electrophoresis system (GE Healthcare) at 15 °C. SDS-PAGE was run at a constant power of 2.5 W/gel for 15 min, and switched to 8 W/gel until the bromophenol blue frontier reached the bottom of the gel. Gels were scanned while still between the glass plates at 480/530 nm (Cy2), 540/595 nm (Cy3), and 635/ 680 nm (Cy5) using a Ettan DIGE Imager (GE Healthcare). Postsilver staining was performed as we previously described.14 DIGE Analysis. Decyder software (version 6.5; GE Healthcare) was used to compare abundance changes. Both the Differential In-gel Analysis (DIA) and the Biological Variation Analysis (BVA) modules were used: the former to codetect and quantify the spots on a given gel in terms of the ratios of the Cy3 and Cy5 sample volumes to the standard Cy2 volume, and the latter to match the spots and standardize the ratios across the gels accounting for the observed differences in the Cy2 sample volumes on the gels. No manual editing of the data was performed. Eighteen protein-spot maps from six gels were matched and average abundance changes were calculated. The Decyder program uses pixel intensity and performs Student’s t-tests and one-way analysis of variance (ANOVA) to determine whether a spot undergoes a statistically significant (P < 0.05) change during transit. The validity of these changes was then confirmed by manual inspection of the gels. To better evaluate the results, the false discovery rate (FDR) and the q-value were also determined by the significance analysis of microarrays (SAM) method (SAM version 3.05). In this analysis, the FDR was set at 10%. In-Gel Digestion. For protein identification, 2-D gels containing unlabeled nuclear samples (400 µg) were run and visualized by silver staining. Spots of interest were chosen from the gels and subjected to in-gel digestion. Briefly, the spots were destained in a freshly prepared 1:1 solution of 30 mM potassium ferricyanide and 100 mM sodium thiosulfate for 1-2 min until the brownish color disappeared. They were then rinsed three times with distilled water to stop the reaction. Next, the gel pieces were incubated in 100 mM ammonium bicarbonate for 5 min, and dehydrated with acetonitrile (ACN) and dried in a Speed-Vac (Thermo Savant). Then, they were further swollen in a digestion buffer containing 40 mM ammonium bicarbonate, 9% ACN and 20 µg/mL trypsin (Sigma, proteomics sequencing grade) in an ice-cold bath. After 30 min, the extra liquid was removed, the same buffer without trypsin was added
Differential Nuclear Proteomes in Response to MNNG Exposure to keep the gels wet, and they were incubated at 37 °C overnight. Peptides were extracted twice with 50% ACN and 5% trifluoroacetic acid (TFA) at room temperature and dried in a vacuum centrifuge. Protein Identification by MS. Peptides were dissolved in 0.1% TFA, and 1 µL of the mixture was mixed with an equal volume of 10 mg/mL CHCA saturated with 50% ACN in 0.1% TFA. The total 2 µL solution was applied onto a MALDI-TOFMS target and allowed to air-dry. The spectra were obtained using a Voyager-DE STR MALDI-TOF mass spectrometer (Applied Biosystems, Foster City, CA) in a reflector/delayed extraction mode. Peptide mass mapping was performed in the positive ion reflector mode, and 300 single-shot spectra were accumulated with a mass range from 800 to 4000 Da. Mass spectra were analyzed by Data Explorer 4.0 (Applied Biosystems) with baseline correction and noise filter/smooth (noise filter correlation: 0.5-0.7). External calibration was carried out using ACTH18-39 and P14R (Sigma) and the internal calibration was performed using the autolytic peaks of trypsin. Proteins were identified using the Mascot software (http:// www.matrixscience.com) and the nrNCBI database, based on peptide mass fingerprinting. Proteins matching more than four peptides and with a MASCOT score higher than 64 were considered statistically significant (P < 0.05). The search parameters were defined as follows: cysteine as carbamidomethylated (fixed modification), methionine as oxidized (variable modification), maximal mass tolerance of 100 ppm and up to 1 missed enzymatic cleavage. Western Blot Analysis. Nuclear and cytoplasmic protein extracts were isolated using the NE-PER kit (Pierce, Rockford, IL) as directed by the manufacturer’s instructions. After SDSPAGE, gels were transferred electrophoretically onto a nitrocellulose membrane (Micron Separations, Inc., Westborough, MA) and blocked for 2 h with Tris-buffered saline (TBS) containing 5% skim milk. Primary antibodies included rabbit anti-human 14-3-3ζ (Santa Cruz, CA) and mouse anti-human HNRPK (Santa Cruz, CA). Membranes were incubated with the primary antibody overnight at 4 °C, washed three times with TBS containing 0.1% Tween-20, incubated with a horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature, and then developed with a chemiluminescence reagent (EZ-ECL Chemiluminescence Detection Kit, Israel). For detecting Ku70 and GAPDH as the loading control, the same blot was first washed several times with TBS containing 0.1% Tween-20 and then stripped by incubation with a stripping buffer containing 65 mM Tris-HCl, pH 7.5, 100 mM β-mercaptoethanol, and 2% SDS for 30 min at 50 °C, followed by washing four times with TBS containing 0.1% Tween-20 at room temperature. Ku70 and GAPDH were then probed by the same steps used to detect 14-3-3 ζ and HNRPK. The values from three independent experiments were compared by using an unpaired Student’s t-test.
Results 2-D DIGE Analysis of Differentially Expressed Nuclear Proteins after MNNG Treatment. Our previous studies showed that low concentrations of MNNG (0.2∼2 µM) with no obvious cytotoxic effect can induce diverse cellular responses, including NTM at undamaged DNA bases, the decrease of DNA replication fidelity, ER stress and the activation of several signal transduction pathways. However, the higher concentrations of MNNG (10∼20 µM) cannot induce these effects.5,6,9-13 Our proteomic analysis of whole cell lysates also indicated that with
research articles
Figure 1. Representative postsilver stained 2D-DIGE gel image indicating differentially expressed nuclear protein spots identified by MS analysis after MNNG exposure. The nuclear protein extracts obtained from FL cells with Cy2/Cy3/Cy5 labeling were separated on a pH 4-7 linear IPG strip, followed by 12.5% SDSPAGE. The gel was poststained by silver staining and differentially expressed proteins were numbered. Circles and rectangles indicate protein spots that were up-regulated or down-regulated by MNNG treatment, respectively.
an increase in the concentrations of MNNG, the proportion of up-regulated proteins decreased, while down-regulated proteins gradually increased, reflecting the enhancement of its cytotoxic effects.14 Therefore, to further understand the active and complicated molecular events triggered by this mutagen, in the present study, the differential nuclear proteomes were analyzed in human amniotic epithelial cells after exposure to low concentrations of MNNG (0.25 and 1 µM) by using 2-D DIGE. Samples were prelabeled with either Cy3 or Cy5 fluorescent dyes, and then each Cy3/Cy5-labeled sample pair was mixed with a Cy2-labeled pooled standard sample containing an equal amount of all 12 samples (three treatment groups including DMSO, 0.25 and 1 µM MNNG, four biological replicates each) before running all three dye-labeled samples together on the same gel. Furthermore, we interchanged Cy3 and Cy5 labeling design to avoid dye labeling-bias. Image analysis revealed that typically 1500 protein spots were detected per gel and the set of 18 gel images was highly reproducible, since over 90% of the spots were detected in all gels. With a P-value of
