ARTICLE pubs.acs.org/jpr
H Ferritin Gene Silencing in a Human Metastatic Melanoma Cell Line: A Proteomic Analysis Maddalena Di Sanzo,§ Marco Gaspari,§ Roberta Misaggi,§ Francesco Romeo,§ Lucia Falbo,§ Carmela De Marco,† Valter Agosti,§ Barbara Quaresima,§ Tullio Barni,§ Giuseppe Viglietto,§ Martin Røssel Larsen,z Giovanni Cuda,§ Francesco Costanzo,*,§ and Maria Concetta Faniello§ §
Dipartimento di Medicina Sperimentale e Clinica “G. Salvatore”, Universita degli Studi di Catanzaro “Magna Græcia”, viale Europa, Campus Universitario, “S. Venuta” - 88100 Catanzaro, Italy † Laboratorio di Oncologia Molecolare, BioGem s.c. a r.l., Ariano Irpino (AV), Italy z Department of Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark
bS Supporting Information ABSTRACT: Ferritin, the major intracellular iron-storage protein, is made of 24 subunits of two types, H and L. Besides regulating intracellular iron homeostasis, it has been found that ferritin, in particular the H subunit (FHC), is involved in different biological events such as cell differentiation and pathologic states (i.e., neurodegeneration and cancer). This study is aimed at investigating the whole-cell proteome of FHC-expressing and sh-RNA-silenced human metastatic melanoma cells (MM07m) in the attempt to identify and classify the highest number of proteins directly or indirectly controlled by the FHC. We identified about 200 differentially expressed proteins and classified them in clusters on the basis of their functions, as proteins involved in metabolic processes, cell adhesion, migration, and proliferation processes. Some of them have captured our attention because of their involvement in metabolic pathways related to tumor progression and metastasis. In vitro assays confirmed that the FHC-silenced MM07m cells are characterized by a decreased growth activity, a reduced invasiveness, and a reduced cell adhesion capability. Moreover, nude mice (CD1 nu/nu), subcutaneously injected with FHC-silenced MM07m cells, showed a remarkable 4-fold reduction of their tumor growth capacity compared to those who received the FHC-unsilenced MM07m counterpart. In conclusion, these data indicate that gene silencing technology, coupled to proteomic analysis, is a powerful tool for a better understanding of H ferritin signaling pathways and lend support to the hypothesis that specific targeting of this gene might be an attractive and potentially effective strategy for the management of metastatic melanoma. KEYWORDS: cancer biology, gene expression, ICAT, siRNA, tandem mass spectrometry, FHC
’ INTRODUCTION Iron is an essential macronutrient for living cells, due to its role in processes such as DNA and adenosine triphosphate synthesis.1 Deregulated iron levels in the cell induce the generation of free radicals and of reactive oxidative species (ROS) through Fenton chemistry, which causes lipid peroxidation, DNA breakages, and other forms of cellular damage. Because of this, iron uptake and storage are among the most highly regulated processes in the cell. The major iron storage molecule, able to keep iron in a soluble and nontoxic form, is a 24-mer protein complex named ferritin. The ferritin shell is composed by the variable array of a heavy (FHC) and light (FLC) subunits; FLC is particularly represented in ferritin molecules extracted from liver and spleen, while FHC is particularly abundant in ferritins from heart, kidney, and different types of neoplastic cells. The two subunits, which share extensive homology in their aminoacid sequence, are functionally r 2011 American Chemical Society
distinct: FHC is responsible for the ferroxidase activity of the ferritin molecule, whereas FLC is mainly involved in iron storage.2 FHC and FLC are coded for by two different genes,3 the activity of which is regulated both at the transcriptional46 and the post-transcriptional levels.7 The coordinated interplay of these control mechanisms accounts for the different expression of the H and L genes among distinct cell types3 during cell differentiation8 and neoplastic transformation,9 as well as in response to some environmental stimuli.9,10 The post-transcriptional control of the ferritin genes is iron-dependent11 and acts mainly on the translational efficiency of the H and L mRNAs.12 The transcription driven by the two ferritin promoters is modulated by many molecules;13,14 in human, the H promoter is positively Received: July 27, 2011 Published: November 01, 2011 5444
dx.doi.org/10.1021/pr200705z | J. Proteome Res. 2011, 10, 5444–5453
Journal of Proteome Research regulated by cAMP15 and by cJun,16 whereas E1A15 and p5317 act as transcriptional repressors. Aside from its central role as antioxidant molecule, an emerging body of evidence identifies new roles for ferritin in physiologic and pathologic processes. For instance, it has recently been demonstrated that ferritin plays a key role in regulating the levels of angiogenesis during inflammation and malignancy,18 in the onset of HIV-mediated neuropathology in patients with history of drug abuse,19 in the protection against the UV-induced DNA damage in corneal epithelial cells,20 and in the pathogenesis of the rare triple A syndrome.21 A survey of the literature in this field indicates that (i) the FHC is involved in the vast majority of these cellular processes and (ii) the FHC may physically interact with several signaling molecules involved in critical cellular pathways, such as the CXCR4 receptor, expressed in a variety of human malignancies,22 and the ALADIN protein (alacrima-acalasia-adrenal insufficiency neurological disorder), whose mutations cause the triple A syndrome.21 In addition, the FHC has been found in the cell nuclei in several cell types and in distinct moments of the cell life.23 To date, the functional role of the FHC in the nucleus has not been determined. In addition to a protective role for the DNA against oxidative and UV damage, a role for FHC in the control of transcription of certain genes, such as that of the β-globin, has also been proposed.2426 In this study we have analyzed the whole-cell proteome of the FHC-silenced MM07m cells, a human skin melanoma cell line established from a supraclavicular lymph node metastasis, in comparison with that of the wild-type MM07m counterpart, characterized by high levels of FHC. The aim was to identify and classify proteins directly or indirectly modulated by the FHC. About 200 differentially regulated proteins were identified and classified in clusters on the basis of their function, i.e., enzymes, proteins involved in cell adhesion, migration and proliferation processes, response to stress factors, etc. Our attention was drawn by a specific group of proteins known to be involved in metabolic processes related to tumor progression and metastasis, which expression was significantly reduced in the FHC-silenced cells. This suggests that the FHC-gene silencing can be accompanied by a reduced potential in terms of proliferation, invasiveness, and adhesion ability of the silenced cells. In vitro assays confirm that the FHC-silenced MM07m cells are characterized by a decreased growth activity, a reduced invasiveness, and a reduced cell adhesion capability. Moreover, the FHC-silenced cells show a remarkable 4-fold reduction of their tumor growth capacity when transplanted in CD1 nu/nu nude mice. In conclusion, the coupling of gene silencing technology with the proteomic analysis appears as an original and very promising approach to identify the molecular patterns in which the silenced gene might be involved.
’ EXPERIMENTAL PROCEDURES
ARTICLE
Preparation of Lentiviral Supernatants and Transduction of MM07m Cells
HEK-293T cells (5 106) were grown on 10-cm plates to 7080% confluence and cotransfected with 10 μg shRNA lentiviral DNA, 2 μg pCMV-VSV-G expressing envelope plasmid, and 18 μg packaging viral CMV δ 8.9 plasmid, using the calcium phosphate precipitation method. Eight hours later, fresh medium was added, and cells were cultured for an additional 2 days. The medium was harvested 48 h post-transfection and filtered through a 0.45 μm filter. The supernatants from HEK293T cultures were used to cross-transduce MM07m cells in the presence of 8 μg/mL Polybrene (Sigma-Aldrich), and subsequently clones were selected by puromycin (1 μg/mL), giving rise to pool 1 (shRNA 29429) and pool 2 (shRNA 29432), respectively.
RNA Extraction and Semiquantitative Reverse Transcriptase Polymerase Chain Reaction (RT-PCR)
Total RNA was extracted from MM07m, shRNA control, pLK0.1 empty vector, shRNA pool 1, and shRNA pool 2 using TRIzol reagent (Invitrogen) and reverse-transcribed with a reverse transcription kit (Invitrogen). cDNA (2 μL) was amplified for the H ferritin gene with the following primers: forward 50 -cat caa ccg cca gat caa c-30 ; reverse, 50 -gat ggc ttt cac ctg ctc at-30 . A human GAPDH cDNA fragment was amplified as the internal control for the amount of cDNA in the PCR. The amplified products were electrophoresed on 1.5% agarose gels. Western Blotting Analysis
Confluent plates of melanoma cells MM07m, shRNA control, pLK0.1 empty vector, shRNA pool 1, and shRNA pool 2 were lysed in buffer [20 mM Hepes pH 7.9, 420 mM NaCl, 1% Triton X-100, 1 mM EDTA, 25% glicerol, 1 mM PMSF, 1 mM Na3VO4, 1 mM DTT, 1 μg/mL aprotin, 1 μg/mL leupeptin] for 30 min on ice. After removal of cell debris by centrifugation (12,000g, 30 min), the concentration of proteins in the supernatant was measured by the Biorad protein assay according to the manufacturer’s instructions (Biorad Laboratories). A total of 70 μg of protein extract was boiled for 10 min in SDS sample buffer and separated by 12% SDS-PAGE, and the proteins were transferred to a nitrocellulose membrane by electroblotting. Nonspecific reactivity was blocked in nonfat dry milk in TPBS [5% (w/v) milk in PBS (pH 7.4) and 0.005% Tween20] for 2 h at room temperature. The membrane was treated with rabbit anti-H ferritin antibody (1:200; Santa Cruz Biotechnology) for 2 h at room temperature, followed by incubation with goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (1:5000; Santa Cruz Biotechnology). The membrane was developed by ECL-Western blot detection reagents according to the manufacturer’s instructions (Santa Cruz Biotechnology). γ-Tubulin was used as a loading control. Proliferation Assay
Cell Culture
MM07m (supraclavicular lymph node metastasis) cells were obtained from a patient of the Surgical Oncology Unit and subsequently tested and authenticated according to the general guidelines.27,28Cells were cultured in Roswell Park Memorial Institute medium (RPMI 1640) (Life Technologies) supplemented with 10% FBS, 10 units/mL penicillin and 10 mg/mL streptomycin at 37 °C in a humidified atmosphere containing 5% CO2.
Melanoma cells (2 103/200 μL) MM07m, shRNA control, pLK0.1 empty vector, shRNA pool 1, and shRNA pool 2 were plated in 96-well microplates. After 24, 48, and 72 h of culture, 20 μL of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) solution (5 mg/mL in PBS) was added, and the cells were incubated at 37 °C for 4 h. After incubation the supernatant was removed, and 150 μL of isopropyl alcohol was added to each well. Absorbance was measured on an ELISA Bio-Rad Microplate Reader using a test wavelength of 492 nm
5445
dx.doi.org/10.1021/pr200705z |J. Proteome Res. 2011, 10, 5444–5453
Journal of Proteome Research and a reference wavelength of 630 nm. Each experiment was performed in triplicate. Invasion Assay
The invasiveness of MM07m, shRNA control, pLK0.1 empty vector, shRNA pool 1, and shRNA pool 2 was assayed using modified Boyden chambers. A polycarbonate filter (pore size, 8 μm) (BD Biosciences) separating the upper and lower compartments was coated with 50 μg of reconstituted basement membrane (Matrigel, BD Biosciences). Cells were washed 3 times with RPMI culture media containing 1% FBS. RPMI/ 1% FBS containing 1.0 105 cells in 100 μL was introduced into the upper compartment; the lower compartment contained 5 μg/mL fibronectin as an attractive substrate in 600 μL of RPMI. After 48 h of incubation at 37 °C, cells that had penetrated the Matrigel were quantified by colorimetric MTT assay. Cells on the upper surface of the filter that had not invaded through the Matrigel were removed completely with cotton swabs. Cells that had invaded remained on the filter, and 25 μL of MTT solution (5 mg/mL in PBS) was added to each well. After 4 h of incubation at 37 °C, the cells on the filter formed dark-blue crystals of MTT formazan. The filter was then moved to another well containing 150 μL of isopropyl alcohol to dissolve the formazan crystals. After 15 min, the solution containing dissolved formazan was poured into a 96-well microplate, and absorbance was measured on a Bio-Rad Microplate Reader using a test wavelength of 492 nm and a reference wavelength of 630 nm. Each experiment was performed in triplicate. Adhesion Assay
Melanoma cells (2 103/200 μL) MM07m, shRNA control, pLK0.1 empty vector, shRNA pool 1, and shRNA pool 2 were plated in 96-well microplates. Culture plates were coated with 50 μL of fibronectin (10 μg/mL) (BD Biosciences) in phosphatebufferd saline overnight at 4 °C. Plates were then blocked with 0.5% bovine serum albumin in RPMI for 4560 min at 37 °C followed by washing three times with 0.1% bovine serum albumin in RPMI. The cells were harvested with trypsin/EDTA, washed twice, and resuspended in RPMI. Cells (2.5 104/well) were added to each well in triplicate and incubated for 30 min at 37 °C. Plates were then washed three times with RPMI to remove unbound cells. Cells remaining attached to the plates were quantified with MTT solution (40 μL per well, 2 mg/mL in PBS) for 4 h at 37 °C. After discarding the supernatant, 200 μL of isopropyl alcohol was added, and the absorbance of the color substrate was measured with a Bio-Rad Microplate Reader using a test wavelength of 492 nm and a reference wavelength of 630 nm. Each experiment was performed in triplicate. Animal Models
All mice were maintained under specific pathogen-free conditions in a Laboratory Animal Facility (BioGem s.c. a r.l., Ariano Irpino Avellino, Italy), and all studies were conducted in accordance with Italian regulations for experiments on animals. Melanoma cells MM07m, shRNA control, pLK0.1 empty vector, shRNA pool 1, and shRNA pool 2 (2 106) were injected s.c. into the flanks of male CD1 nu/nu mice in a final volume of 100 μL. Eight mice for each experimental point were used. The size of the transplanted tumors was measured every 7 days and the tumor volume was calculated using the formula V = 1/2(L 3 W2). After 35 days, animals were sacrificed, and tumors were removed and processed for RNA extraction.
ARTICLE
Protein Digestion and ICAT Labeling
Cell pellets were lysed in 100 mM Tris (pH 8.5), 250 mM NaCl, and 0.5% SDS for 30 min on ice. After removal of cell debris by centrifugation (12,000g, 30 min), the concentration of proteins in the supernatant was measured by BCA protein assay. Samples were diluted in lysis buffer in order to have a protein concentration of 2 μg/μL. Proteins were processed according to the protocols supplied in the Cleavable ICAT Reagent Kit for protein Labeling (AB Sciex, Foster City, CA) adjusted to take into account a higher starting SDS concentration. Briefly, 80 μL (160 μg) of melanoma cells MM07m lysates and shRNA lysates (pool 2, shRNA 29432) were reduced by 2.5 mM Tris (2-carboxyethyl) phosphine (TCEP) for 10 min at 100 °C. Proteins were labeled with either biotin-conjugated (13C) (heavy) ICAT reagent (H-label) or with the biotin-conjugated (12C) (light) ICAT reagent (L-label). Samples were mixed in pairs and digested with endoproteinase Lys-C (Sigma, St. Louis, MO) overnight at 37 °C. Subsequently, samples were diluted with H2O to a final volume of 400 μL in order to have an SDS concentration of 0.1%. Finally, 10 μg of trypsin (Sigma, St. Louis, MO) was added ,and enzymatic digestion was allowed to proceed for an additional 8 h at 37 °C. Peptide mixtures were purified by cation exchange chromatography and purified by avidin affinity chromatography, and finally the biotin tag was cleaved off according to the manufacturer’s instructions. A second 80 μL aliquot of MM07m and shRNA cell lysates was processed in parallel, following the same experimental procedures described above, with the exception that for the ICAT labeling step MM07m lysate was labeled with the ICAT Light reagent, whereas shRNA lysate was labeled with the ICAT Heavy Reagent. NanoLCMS/MS Analysis
Nanoscale chromatography (nanoLC) was performed on an Easy nano LC system from Proxeon (Odense, Denmark). Lyophilized ICAT-labeled samples were reconstituted in 50 μL of mobile phase A (see below). A 5 μL aliquot of the reconstituted samples was loaded onto a 100 μm i.d. 360 μm outer diameter capillary tip in-house packed with ReproSil Pur C18 AQ 3 μm (Dr. Maisch, Ammerbuch-Entringen, Germany) reversed phase material. Column length was 20 cm. The peptides were eluted using a gradient from 100% phase A (0.1% formic acid) to 40% phase B (0.1% formic acid, 80% acetonitrile) over 180 min at 300 nL/min directly into a LTQ-Orbitrap XL mass spectrometer (Thermo Scientific, San Jose CA) for tandem mass spectrometry (MS/MS) analysis. Spray voltage was set to 1800 V (positive ion mode). The capillary column was washed with 100% phase B for 10 min, and subsequently equilibrated for 20 min at 100% phase A. The LTQ-Orbitrap XL was operated in a data-independent mode automatically switching between MS and MS/MS of the 7 most intense ions from the MS scan. The signal intensity threshold for selection for MS/MS was set to 30000 NL, the activation time was 30 ms and previously selected peptides were excluded for 20 s. Duplicate nanoLCMS/MS analyses were performed for both ICAT mixtures. Data Analysis
Raw datafiles were processed using Proteome Discoverer software (Thermo Scientific), version 1.2.0.207. (beta version) using default parameters. The processed data were searched on an in-house Mascot search engine server (version 2.2), against the human SwissProt annotated database (version 2010_09, 20359 entries, released on August 10, 2010) using the following 5446
dx.doi.org/10.1021/pr200705z |J. Proteome Res. 2011, 10, 5444–5453
Journal of Proteome Research
ARTICLE
parameters: initial MS tolerance, 5 ppm; MS/MS tolerance, 0.5 Da; dynamic modifications, Oxidation (M), ICAT-C:13C(9) (C), ICAT-C (C); enzyme, trypsin; max missed cleavages, 2; quantitation, ICAT. False-discovery rate was set to 1% in Proteome Discoverer using a decoy database search. Mascot score cutoff value was 14. The complete list of identified and quantified peptides and corresponding proteins in experiment 1 (shRNA = heavy; MM07m = light) together with parameters for processing and searching MS data is available as Supporting Information 1. Supporting Information 2 reports data for the reversed labeling experiment (shRNA = light; MM07m = heavy). Statistical Analysis
Data analysis for in vitro and in vivo assays, was done by Student t test assuming equal variances. Data are the mean ( SEM of three independent experiments, and each experiment includes triplicate sets.
’ RESULTS H Ferritin Gene Silencing in MM07m Cells
In the MM07m cells the gene coding for the ferritin heavy chain is consistently up-regulated both at the mRNA and protein levels, in comparison with the levels found in their primary counterpart as well as in melanocitic cells and in several melanoma cell lines (data not shown). Thus, the MM07m cells represent a good model system to investigate the functional role of FHC by means of selective gene silencing. To this aim, the MM07m cells were stably transduced with a lentiviral DNA containing either shRNAs that specifically target the H ferritin mRNA (29429 and 29432) or an shRNA without any significant homology to known human mRNAs. As a control, the cells were also transduced with the empty vector. Transfected clones were then selected with puromycin. The H ferritin mRNA and protein levels were analyzed in the selected clones by reverse transcription-PCR and Western blot analysis. As shown in panels A and B of Figure 1, the endogenous expression of FHC is consistently reduced in the cell pools 1 and 2, transfected with the 29429 and 29432 shRNAs, respectively. ICAT Labeling and Mass Spectrometry Analysis of H Ferritin Silenced versus Unsilenced MM07m Cells
Protein extracts from the FHC-silenced (pool 2, shRNA 29432) and unsilenced cells were labeled with ICAT heavy and ICAT light reagents, respectively. An experiment with inverted labeling (ICAT light-labeled silenced cells and ICAT heavylabeled unsilenced cells) was performed in parallel. Following ICAT labeling, sample mixing in pairs, tryptic digestion, and isolation of cysteine containing peptides by affinity chromatography, the simplified peptide mixture was subjected to nanoLCMS/MS analysis. A total of 692 and 510 distinct proteins were identified and quantified, respectively, across experiment 1 and 2 (Supporting Information 1 and 2). We have previously shown that technical variation in our ICAT-based experiments allows the detection of protein abundant changes >33% .29 In this context, protein abundance fold changes of >2 or 2), whereas 213 proteins were found downregulated in silenced cells (H:L ratio 2), whereas 152 proteins were found downregulated in unsilenced cells (H:L ratio
