If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Life Science, Institute of Molecular and Cellular Biology, National Taiwan University, Taipei, TaiwanGraduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taipei, TaiwanThe Research Center of Developmental Biology and Regenerated Medicine, National Taiwan University, Taipei, TaiwanCenter for Computational and Systems Biology, National Taiwan University, Taipei, Taiwan
Mesenchymal stem cells (MSCs) enhance lung cancer development through extracellular factor secretion.
•
Phosphoproteomics discover MSCs-regulated phosphosignaling in the lung cancer.
•
Ectopic ATP synthase on MSCs surface produces ATP into the tumor microenvironment.
•
MSC-secreted extracellular ATP mediates the phosphorylation of the ERK/c-Fos axis.
Abstract
The tumor microenvironment (TME), which comprises cellular and noncellular components, is involved in the complex process of cancer development. Emerging evidence suggests that mesenchymal stem cells (MSCs), one of the vital regulators of the TME, foster tumor progression through paracrine secretion. However, the comprehensive phosphosignaling pathways that are mediated by MSC-secreting factors have not yet been fully established. In this study, we attempt to dissect the MSC-triggered mechanism in lung cancer using quantitative phosphoproteomics. A total of 1958 phosphorylation sites are identified in lung cancer cells stimulated with MSC-conditioned medium. Integrative analysis of the identified phosphoproteins and predicted kinases demonstrates that MSC-conditioned medium functionally promotes the proliferation and migration of lung cancer via the ERK/phospho-c-Fos-S374 pathway. Recent studies have reported that extracellular ATP accumulates in the TME and stimulates the P2X7R on the cancer cell membrane via purinergic signaling. We observe that ectopic ATP synthase is located on the surface of MSCs and excreted extracellular ATP into the lung cancer microenvironment to trigger the ERK/phospho-c-Fos-S374 pathway, which is consistent with these previous findings. Our results suggest that ectopic ATP synthase on the surface of MSCs releases extracellular ATP into the TME, which promotes cancer progression via activation of the ERK/phospho-c-Fos-S374 pathway.
). In addition to the cellular components, noncellular components secreted from tumor or stromal cells, such as growth factors, chemokines, cytokines, and extracellular matrix proteins, also play essential roles in the TME (
). Tumor-associated stromal cells, which surround the malignant tumor tissue, include cell types such as stromal fibroblasts, endothelial cells, macrophages, immune cells, and mesenchymal stem cells (MSCs). MSCs are considered multipotent stromal cells that have the potential to differentiate into multiple cell lineages, including osteoblasts, chondrocytes, and adipocytes (
). An increasing number of studies have reported that MSCs could be attracted to the tumor site and cause tumor cells to undergo functional modulation via paracrine factors secreted by MSCs or via direct interaction with tumor cells (
Mesenchymal stem cells isolated from human gliomas increase proliferation and maintain stemness of glioma stem cells through the IL-6/gp130/STAT3 pathway.
). Thus, understanding the intercellular communication network between tumor cells and MSCs within the tumor stroma may provide a new avenue for the development of anticancer therapies.
Recent investigations into ectopic adenosine 5′-triphosphate (ATP) synthase (ecto-ATP synthase) highlight the importance of this cell-surface protein in many diseases. Ecto-ATP synthase is an enzyme complex that has been observed on the outer surface of the plasma membrane of various cell types (
). The F1 region consists of three repetitions of subunits α and β and one repetition of subunits γ, δ, and ε (α3β3γδε). The Fo region contains subunits a, b, d, F6, oligomycin sensitivity–conferring protein, and a transmembrane c-ring (
). From a functional point of view, ecto-ATP synthase retains the catalytic activity of ATP synthase to phosphorylate ADP and generate extracellular ATP (eATP) (
). Accruing evidence seems to suggest that the concentration of eATP in tumor tissue is vital for the determination of cell growth, invasion, and even chemoresistance (
). Therefore, clarification of the role of ecto-ATP synthase in the TME may give new insight into the development of therapeutic strategies for cancer.
Recent cancer research suggests that P2X purinergic receptors (P2XRs) act as the specific plasma membrane receptors for eATP (
). In response to stimulation with eATP, P2XRs trigger intracellular signal transduction cascades to control multiple physiological and pathophysiological processes (
). Among the P2XR family, the P2X purinergic receptor 7 (P2X7R) subtype is commonly overexpressed in most malignant tumor cells, including those of primary breast cancers, prostate cancers, melanoma, colon cancers, neuroblastoma, and non–small cell lung carcinoma (
). The stimulation of P2X7Rs, located on the cell surfaces of cancer cells, by eATP in the TME activates either the PI3K-AKT pathway or the ERK1/2-MAPK pathway, both of which are involved in the regulation of tumor growth (
Basal activation of the P2X7 ATP receptor elevates mitochondrial calcium and potential, increases cellular ATP levels, and promotes serum-independent growth.
P2X7 receptor antagonism inhibits p38 mitogen-activated protein kinase activation and ameliorates neuronal apoptosis after subarachnoid hemorrhage in rats.
). In addition to mediating the P2X7R-dependent signaling pathway, the stimulation of P2X7R also enhances the release of cytokines and matrix metalloproteinases to promote tumor cell motility, migration, and invasion (
). In this study, MSC-culture medium was collected and used to treat lung cancer cells in order to analyze the tumor functions that are regulated by the paracrine factors secreted by MSCs. A time-series phosphoproteomic analysis was performed to further improve our understanding of the molecular mechanisms triggered by MSCs in cancer. These results may provide new insights into the TME and thus be useful for improving cancer therapy.
Experimental Procedures
Cell Culture
MSCs expressing red fluorescent protein (hTERT/RFP-cbMSCs, abbreviated as MSCs hereafter) (
Establishment of immortalized mesenchymal stromal cells with red fluorescence protein expression for in vivo transplantation and tracing in the rat model with traumatic brain injury.
) and A549 lung cancer cells were purchased from the Bioresource Collection and Research Center (Food Industry Research and Development Institute, Hsinchu, Taiwan). Low-motility cells (LM cells) were selected via a Transwell migration assay from the A549 cell line, as per the methods outlined in a previous study (
). MSCs were cultured in minimum essential medium (MEM, Gibco Laboratories) supplemented with 20% fetal bovine serum (FBS; Gibco Laboratories), 4 ng/ml human fibroblast growth factor-basic, 1 mM sodium pyruvate (Thermo Fisher Scientific), 30 μg/ml hygromycin (Millipore), and 200 μg/ml geneticin (Millipore). A549, CL1-0, and LM cells were grown in Dulbecco’s modified Eagle’s medium (DMEM; Gibco Laboratories) supplemented with 10% FBS. All cell lines were cultured in an incubator at 37 °C and with 5% CO2. Cell lines used in this study were authenticated by short tandem repeat profiling and tested for Mycoplasma contamination.
Conditioned Medium Collection
MSCs (1 × 106) were plated in 10-cm culture dishes in MEM containing 20% FBS. After culturing for 72 h, the culture medium from the MSCs (MSC-CM) was collected and then centrifuged for 10 min at 2000g to remove cell debris. The supernatant was stored at −80 °C. The culture medium from the same number of A549 or LM cells was collected as Control-CM. The culture medium from MSCs or lung cancer cells treated with ecto-ATPi, citreoviridin, for 72 h was collected as ecto-ATPi-Control-CM or ecto-ATPi-MSC-CM.
Antibodies and Reagents
The antibodies used in this study were as follows: anti-phospho-c-Fos (pSer374) (Santa Cruz Biotechnology), anti-c-Fos (Cell Signaling Technology), anti-phospho-ERK1/2 (pThr202/pTyr204; Cell Signaling Technology), anti-ERK1/2 (Cell Signaling Technology), anti-HA (Biolegend), anti-GAPDH (Millipore), anti-β-actin (Millipore), horseradish peroxidase–conjugated anti-rabbit IgG H&L (Abcam), and horseradish peroxidase–conjugated anti-mouse IgG H&L (Abcam). ERK1/2 inhibitor (SCH772984; TargetMol) and competitive P2X7R antagonist (A 438079; Abcam) were dissolved in DMSO (Sigma–Aldrich) and stored at −20 °C.
Cell Proliferation Analysis
Cells were seeded onto six-well plates at 4 × 104 cells per well. After attachment for 24 h, the culture medium was replaced with serum-free DMEM for 24 h. The cells were then incubated with MSC-CM for the indicated time points. Cell viabilities were identified through 0.4% trypan blue solution (VWR), and the total number of viable cells for each well was measured using a hemocytometer. Cells incubated with Control-CM served as control groups.
MTS Assay
Cells were seeded onto 96-well plates at 1 × 103 cells per well. At the indicated time points, cell viability was detected in the MSC-CM and Control-CM groups using CellTiter 96 AQueous MTS Reagent (Promega Corporation) according to the manufacturer’s protocol. The relative cell viability and raw absorbance values are presented as histograms.
Migration Assay
Cell migration was assessed using a Boyden chamber assay involving an 8-μm pore polycarbonate membrane (Corning Incorporated). Cells (3.5 × 104) were seeded onto six-well plates and stimulated with MSC-CM for 72 h, as per the aforementioned procedure. The pretreated cells were then trypsinized and placed in the plate well inserts at a density of 5 × 104 cells per chamber, and conditioned medium was added into the lower Transwell chambers. After incubation for 10 h, the insert membranes were fixed with 3.7% paraformaldehyde at room temperature for 5 min, followed by 100% MeOH overnight. The cells on the top surface of the inserts were wiped off with cotton swabs, and the migrating cells on the other side of the inserts were stained with 10% Giemsa stain (Sigma–Aldrich) for 1 h. After washing with PBS, the stained migrating cells were observed and counted under a microscope. Five fields were randomly selected for each insert membrane, and the average number of migrating cells in each Transwell chamber was represented in bar charts.
Colony Forming Assay
Cells were seeded onto six-well plates at 1 × 103 cells per well and adhered for 24 h. Cells were then incubated with serum-free DMEM for 24 h before being treated with MSC-CM for 14 days for colony formation (fresh MSC-CM was exchanged every 3 days). Colonies were fixed with 100% methanol and stained with 10% Giemsa stain. The colony numbers were quantified using ImageJ (National Institutes of Health).
Experimental Design and Statistical Rationale
LM cells were treated with MSC-CM or with Control-CM for 5 min, 10 min, 30 min, and 60 min. The Control-CM-treated cells served as controls. Two biological replicates of the Control-CM-treated and the MSC-CM-treated LM cells for each time point were obtained, and each biological sample underwent duplicate nanoLC-MS/MS analyses. Phosphosites with differential phosphorylation levels were assessed using ANOVA analysis.
Protein Extraction
The Control-CM-treated cells and the MSC-CM-treated cells were lysed by using PTS lysis buffer composed of 12 mM sodium deoxycholate (Sigma–Aldrich), 12 mM sodium lauroyl sarcosinate (MP Biomedicals), and 50 mM triethylammonium bicarbonate (TEAB; Sigma–Aldrich) (
). Protease inhibitor cocktail (BioShop), phosphatase inhibitor cocktail I (tyrosine phosphatase inhibitors; BioShop), and phosphatase inhibitor cocktail II (serine/threonine phosphatase inhibitors; BioShop) were added into the PTS lysis buffer for the protection of proteins and phosphoproteins. Proteins were extracted using an ultrasonic homogenizer (LABSONIC M ultrasonic homogenizer, Sartorius AG) operating at 60% amplitude and 0.6 cycle duration for 2 min. After centrifugation at 16,000g for 20 min, the supernatants were transferred into new Eppendorf tubes. The concentrations of the proteins of interest were detected using a Pierce BCA Protein Assay Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions.
Sample Preparation for Quantitative Phosphoproteomics
Five hundred micrograms of each protein sample was reduced with 10 mM dithiothreitol (DTT; BioShop) at room temperature for 30 min, and cysteines were carbidomethylated with 55 mM iodoacetamide (Sigma–Aldrich) in the dark at room temperature for 30 min. Proteins were digested with Lys-C (1:100 w/w; WAKO) for 2 h, followed by digestion with MS-grade trypsin (1:100 w/w; Thermo Fisher Scientific) for 16 h at 37 °C. An immiscible organic solvent, ethyl acetate (Sigma–Aldrich), was mixed into the digested peptide samples for the purposes of detergent removal (
). One hundred micrograms of peptide was dried under vacuum and then dissolved in 100 μl of 100 mM TEAB solution and labeled. The peptides obtained from both the Control-CM-treated group and the MSC-CM-treated group for each time point were labeled with 4% formaldehyde-H2 (Sigma–Aldrich) and 4% formaldehyde-D2 (Sigma–Aldrich), respectively. Freshly prepared 0.6 M sodium cyanoborohydride (Sigma–Aldrich) was added as a catalyst. Cold 1% (v/v) ammonium hydroxide (WAKO) was added to stop the dimethyl labeling reaction, and 10% (v/v) trifluoroacetic acid (TFA; Sigma–Aldrich) was used to acidify the isotope-labeled peptides to pH < 3 (
). Finally, an equal volume of H2-labeled peptides (Control-CM group) and D2-labeled peptides (MSC-CM group) were combined and then desalted with homemade SDB-XC StageTips. The phosphopeptides were enriched by hydroxy acid–modified metal oxide chromatography (MOC) using homemade MOC tips that were prepared by packing 1 mg of TiO2 beads (GL Sciences) into 10-μl C8 StageTips (
). The isotope-labeled peptide mixture was combined with an equal volume of solution A (0.1% TFA, 80% acetonitrile [Thermo Fisher Scientific] and 300 mg/ml lactic acid [WAKO]) and then loaded onto MOC tips (100 μg peptide/tip) that were equilibrated with solution A after being preconditioned with solution B (0.1% TFA and 80% acetonitrile). After washing with solution A and solution B, the phosphopeptides were eluted with 0.5% and 5% piperidine (Sigma–Aldrich), acidified, and desalted. Each sample was dried with a centrifugal evaporator before being analyzed by nanoLC-MS/MS.
NanoLC-MS/MS Analysis
LTQ-Orbitrap XL (Thermo Fisher Scientific) coupled with a nanoACQUITY UPLC system (Waters) was used to perform the NanoLC-MS/MS analysis. In brief, samples were loaded into a 2-cm × 180-μm capillary trap column and then separated in a C18 BEH nanoACQUITY column with an internal diameter (I.D.) of 75 μm and a length of 25 cm, at a flow rate of 300 nl/min, where mobile phases were the mixture of buffer A (0.1% formic acid [WAKO]) and buffer B (0.1% formic acid and 80% acetonitrile). Buffer A linear gradient of 10 to 40% buffer B in 90 min and later 40 to 85% buffer B in 10 min was applied throughout this study. Mass spectra from a full survey scan were acquired using the Orbitrap (m/z 300–1500). The resolution of the instrument was set to 60,000 at m/z 400 with an automated gain control value of 106. The top 10 most intense precursor ions were selected from the MS scan for subsequent collision-induced dissociation MS/MS scanning by ion trap (with an automated gain control target of 7000).
Identification, Quantification, and Analysis of Mass Spectrometric Data
Raw MS/MS spectrums were analyzed using MaxQuant software version 1.5.2.8 (http://maxquant.org/) following parameters used in previous studies (
): trypsin specificity allowing for up to two missed cleavage sites; carbamidomethylation on cysteine (C) was set as fixed modification; oxidation on methionine (M) and phosphorylation on serine, threonine, or tyrosine residues (STY) were selected as variable modifications. The minimal peptide length was set at seven amino acids. The first search peptide tolerance was 20 ppm, whereas the main search peptide tolerance was 4.5 ppm. Fourier transform-based mass spectrometry MS/MS match tolerance was set as 20 ppm. Peptide identification was performed using the Andromeda search engine against the 20,386 UniProtKB/SwissProt human protein entries published on Jun, 02, 2021 (
). A false discovery rate of 1%, searching against a randomized target-decoy database created by MaxQuant, was applied for peptide, protein, and site identifications. The intensity value of dimethyl-labeled phosphopeptides was obtained by summing up the eXtracted Ion Current (XIC) in two technical replicates. The normalized ratios of the heavy-light dimethyl-labeled partners (MSC-CM treatment versus Control-CM treatment) were reported by MaxQuant using the robust linear regression of the 2D centroid intensities of isotope pairs and adjusting the median of logarithmized ratios to 0 (
). The normalized ratios were estimated in two biological replicates. A site localization probability of 0.75 was set as the threshold for localization of phosphorylation sites. The significantly regulated phosphosites between the MSC-CM and the Control-CM groups were determined by ANOVA with a p-value cutoff 0.05 in the Perseus environment using default parameters (version 1.6.1.3) (
). To discover the kinase activities in response to MSC-CM treatment, a literature-based kinase-substrate library, Kinase Enrichment Analysis (KEA2), was applied with the default setting (
). Benjamini–Hochberg false discovery rate was used to adjust the p-values for multiple test correction. The predicted kinases and their substrates (only MSC-CM-regulated phosphosites were shown) were employed to construct the kinase-substrate network, which was visualized in Cytoscape (version 3.7.1). All edges represent the known kinase–substrate interactions in the literature.
Plasmid Construction and Transfection
The open reading frame (ORF) region of the human FOS gene (NM_005252.4) was amplified by PCR using Phusion High-Fidelity DNA Polymerase (Thermo Fisher Scientific). Primers for amplifying wildtype c-Fos and mutant c-Fos (c-Fos-S374A) sequences were as follows: forward, 5′-ATGGAGGCCCGAATTCCCACCATGATGTTCTCGGGCTTC; reverse, 5′-CTGGATCCCCGCGGCCGCTCACAGGGCCAGCAGCGTGGGTGAGCTGAGCG (wildtype c-Fos); forward, 5′-ATGGAGGCCCGAATTCCCACCATGATGTTCTCGGGCTTC; reverse, 5′-CTGG ATCCCCGCGGCCGCTCACAGGGCCAGCAGCGTGGGTGCGCTGAGCG (mutant c-Fos). PCR products were purified with Zymoclean Gel DNA Recovery Kit (Zymo Research), cloned into a linear pCMV-HA-N vector (Takara Bio), and digested by EcoRI and NotI (FastDigest; Thermo Fisher Scientific). The mutant c-Fos (c-Fos-S374A) was generated through site-directed mutagenesis using wildtype c-Fos. Constructions were transfected into cells with jetPRIME in vitro DNA and siRNA transfection reagent (Polyplus-transfection SA) according to the manufacturer’s instructions. After incubation for 24 h, the transfected cells were subjected to further experimentation.
Fluorescent Immunocytochemistry
MSCs were seeded at a density of 2.5 × 104 cells per well onto coverslips placed in 12-well culture dishes. MitoTracker (200 nM, Thermo Fisher Scientific) was used for labeling mitochondria in both the permeabilized and nonpermeabilized groups. After fixation with 3.7% paraformaldehyde (Sigma–Aldrich), cells on the coverslips were blocked in 5% bovine serum albumin (BioShop) in Tris-buffered saline (TBS; 50 mM Tris, 150 mM NaCl, pH 7.2–7.4) with or without 0.1% Triton X-100 (Sigma–Aldrich). Cells were then hybridized with antibodies against ATP synthase complex (Abcam) or IgG (Abcam) overnight at 4 °C. IgG antibody was used for excluding the nonspecific signals. After washing with TBS to eliminate nonspecific binding, the coverslips were incubated with Alexa Flour 488–conjugated goat anti-mouse IgG secondary antibody (Abcam) and mounted onto the glass slides with the blue DNA stain DAPI (Thermo Fisher Scientific). Images were taken using a Zeiss LSM 780 laser scanning confocal microscope and adjusted with Zen 2.6 software (blue edition).
Flow Cytometry
MSCs (1 × 106) were seeded onto 10-cm culture dishes and incubated for 72 h. Cells were detached with 1 mM EDTA (J.T. Baker) and resuspended with MEM containing 20% FBS. After fixation, 1 × 106 MSCs were incubated with anti-ATP synthase complex immunocapture antibody [12F4AD8AF8] (ab109867, Abcam) (
A functionally active human F1F0 ATPase can be purified by immunocapture from heart tissue and fibroblast cell lines. Subunit structure and activity studies.
), which is able to detect the native folded whole enzyme, or mouse anti-IgG antibody (ab18457, Abcam) overnight. The cells were then hybridized with Alexa Fluor 488–conjugated goat anti-mouse IgG secondary antibody (Abcam) by rotation for 2 h at room temperature. Triton X-100 (0.1%) in PBS was used for permeabilization. The fluorescence signal was detected using a BD FACSCanto II (BD Biosciences).
ATP Bioluminescence Assay
MSCs were seeded at a density of 2.5 × 104 cells per well in 12-well plates and cultured for 72 h. Cells were pretreated with or without citreoviridin (2 μM), oligomycin (5 μg/ml), carbonyl cyanide 3-chlorophenylhydrazone (1 μM), or P1P5-Di(adenosine-5′) pentaphosphate pentasodium salt (Ap5A) (300 μM) at 37 °C for 5 min. ATP production was started by adding 100 μM ADP alone or 100 μM ADP and 5 mM Pi. After 3 min incubation, the supernatants were collected and centrifuged at 300g for 15 s to stop the reaction. The cell debris was removed by further centrifugation at 2000g for 10 min. An ATP Bioluminescent Assay Kit (Sigma–Aldrich) was used to quantify eATP concentrations, as per the manufacturer’s instructions. In brief, 100 μl of ATP Assay Mix solution, dissolved in sterile water and adjusted to pH 7.8, was added into each well of the 96-well opaque plate and incubated at room temperature for 3 min to hydrolyze endogenous ATP. The same volume of ATP standards or samples was then added into the wells. After gently swirling, the light was immediately measured by FlexStation three microplate readers (Molecular Devices). A standard curve was prepared by ATP in the range of 8 × 10−8 to 2 × 10−4 M to determine the conditional medium's ATP concentration. The concentration of eATP in the conditioned medium was normalized according to the cell number.
Statistical Analysis
Data are displayed as means ± SD (n = 3). The statistical significance of effects was analyzed using Student’s unpaired two-tailed t test for two-group comparisons. A p-value < 0.05 was considered statistically significant.
Results
MSCs Accelerate Lung Cancer Progression
An increasing number of studies have demonstrated that MSCs possess either protumorigenic or antitumorigenic effects in different types of cancers (
). Therefore, we first investigated the role of MSCs in non–small cell lung cancers. Conditioned medium from MSCs (MSC-CM) was collected and used to treat A549 and LM lung cancer cell lines. The results showed that MSC-CM increased the proliferation rate of lung cancer cells (Fig. 1, A and B). An MTS assay showed that cell viability was consistently higher in lung cancer cells treated with MSC-CM (Fig. 1, C and D). The colony-forming ability of the cells was also enhanced in response to MSC-CM treatment (Fig. 1, E and F). In addition, we investigated whether MSC-CM affected the motility of the cells. In both A549 and LM cells, MSC-CM promoted their Transwell migration ability relative to the Control-CM treatment (Fig. 1, G and H). Here, our results suggest that MSCs enhance proliferation and migration in lung cancer cells through paracrine factors.
Fig. 1MSC-CM promotes cell proliferation and migration in lung cancer cells.A and B, numbers of lung cancer cells treated with Control-CM or MSC-CM at the indicated time points were counted using a hemocytometer. C and D, cell viability in Control-CM–treated and MSC-CM–treated groups was detected using an MTS assay. The raw 490 nm absorbance values and the relative cell viability percentages are expressed as bar graphs. E and F, the colony-forming ability of LM and A549 cells treated with or without MSC-CM for 14 days was assessed and is expressed as a percentage relative to the control group. G and H, a Transwell assay was performed to determine the effect of MSC-CM on tumor cell migration. The average number of migrating cells in each Transwell insert was calculated from five randomly selected fields. Representative images from three independent biological replicates are shown in E-H, and the quantitative results are displayed as mean ± standard deviation. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, two-tailed t test. CM, conditioned medium; LM, low-motility; MSC, mesenchymal stem cell.
Quantitative Phosphoproteomic Profiling Reveals the Temporal Dynamics Behind MSC-Regulated Phosphorylation
After verifying the functional effects of MSC-CM in lung cancer cells, we were curious about the molecular mechanisms induced by MSCs. Hence, we performed quantitative phosphoproteomic profiling, which provides a more holistic insight into protein phosphorylation modification, to explore the signaling pathways triggered by MSC-CM. The proteins of LM cells treated with MSC-CM or Control-CM for 5, 10, 30, or 60 min were extracted and digested. Peptides from the control and treatment groups were labeled with formaldehyde-H2 or formaldehyde-D2, respectively. The phosphopeptides were further enriched with TiO2, identified and quantified via nanoscale liquid chromatography-tandem mass spectrometry (nanoLC-MS/MS) and analyzed using a bioinformatics approach (Fig. 2A). We identified a total of 1321 peptides that contained 1958 phosphorylation sites located on 706 proteins, with 97% phosphopeptide enrichment efficiency (Fig. 2B and supplemental Table S1). Most of the identified phosphopeptides were singly or doubly phosphorylated (Fig. 2C). Among all of the identified phosphosites, 78% were distributed on serine (pSer), 19% on threonine (pThr), and 3% on tyrosine (pTyr) (Fig. 2D). ANOVA was then used to identify the significantly regulated phosphosites in response to MSC-CM treatment (supplemental Table S2). We then became curious about the kinases that were responsible for these differential phosphosites. Ten predicted kinases (red rectangles) were enriched and used to construct a kinase–substrate network (Fig. 2E and supplemental Table S3). Among the MSC-mediated phosphosites, the phosphorylation of the proto-oncogene, c-Fos, at serine 374 was approximately doubled in MSC-CM–treated cells at all four time points. Our results imply that the phosphorylation of c-Fos at Ser374 may play a critical role in MSC-triggered tumor progression.
Fig. 2Identification of phosphorylated proteins in lung cancer cells in response to MSC-CM via quantitative phosphoproteomic analysis.A, the overall workflow for establishing the phosphoproteomic profile of LM cells in response to short-term MSC-CM treatment. T represents the MSC-CM–treated group, and C represents the Control-CM–treated group. B, the total number of identified phosphosites, phosphopeptides, and phosphoproteins across all time points. C, the number of phosphopeptides with singly, doubly, triply, quadruply, and quintuply phosphorylated residues. D, the percentages of phosphorylated serine (phoSerine), threonine (phoThreonine), and tyrosine (phoTyrosine) in all identified phosphosites. E, the kinases associated with the phosphorylated sites that were significantly upregulated after being treated with MSC-CM were predicted and are presented as a kinase–substrate network. The color of squares indicates the relative levels of phosphoproteins (MSC-CM treatment/Control-CM treatment). CM, conditioned medium; LM, low-motility; MSC, mesenchymal stem cell.
Phosphorylation of c-fos at Serine 374 Plays a Critical Role in MSC-Induced Tumor Progression
Western blotting further verified that the phosphorylation levels at Ser374 in c-Fos were indeed upregulated in MSC-CM–treated cells compared with the control group (Fig. 3, A–D). To determine whether phosphorylation of Ser374 in c-Fos was involved in functions regulated via MSC-CM, a nonphosphorylatable mutation of c-Fos was generated by site-directed mutagenesis of Ser374 to Ala374 in c-Fos. Lung cancer cells were then transfected with either HA-tagged wildtype c-Fos (HA-c-Fos-WT) or a nonphosphorylatable mutation on Ser374 of c-Fos (HA-c-Fos-S374A). Upon treatment of the cells with MSC-CM, the phosphorylation levels of c-Fos at Ser374 were diminished in HA-c-Fos-S374A–transfected cells relative to HA-c-Fos-WT–transfected cells (Fig. 3E). We further assessed the relevance of phosphorylated c-Fos in MSC-CM–triggered functions. In the presence of MSC-CM, both proliferation rate (Fig. 3F) and migration (Fig. 3, G and H) were enhanced in HA-c-Fos-WT–expressing cells, while this was not the case for HA-c-Fos-S374A–expressing cells. These results indicate that both MSC-mediated tumor cell proliferation and cell migration require the phosphorylation of c-Fos at Ser374.
Fig. 3MSC-CM enhances tumor proliferation and metastasis by increasing phosphorylation of c-Fos at serine 374.A–D, LM cells (A) and A549 cells (B) were treated with Control-CM or MSC-CM at the indicated time points. The total expression levels of c-Fos and phospho-c-Fos-S374 were detected with Western blotting. GAPDH expression levels served as the loading control. The fold changes in phospho-c-Fos-S374 expression between Control-CM–treated and MSC-CM–treated groups were quantified using ImageJ and normalized according to the expression levels of total c-Fos. The quantified results from three biological replicates are displayed as dot plots (C and D). E, cells were transfected with HA-c-Fos-wildtype plasmid or HA-c-Fos-S374A mutation plasmid for 24 h before being treated with MSC-CM. The levels of c-Fos and phospho-c-Fos-S374 were verified using Western blotting and quantified using ImageJ. F, the number of HA-c-Fos-WT and HA-c-Fos-S374A-transfected cells stimulated with or without MSC-CM for 24, 48, 72, or 96 h was counted using a hemocytometer. G and H, HA-c-Fos-WT and HA-c-Fos-S374A-overexpressing cells were treated with Control-CM and MSC-CM, respectively. Their migrating ability was assessed using a Transwell assay. Representative images from three independent biological replicates are shown (G), and the quantitative results are displayed as a histogram (H). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, two-tailed t test. CM, conditioned medium; LM, low-motility; MSC, mesenchymal stem cell.
Factors Secreted by MSCs Trigger the Phosphorylation of ERK1/2 to Promote Tumor Progression
As shown in the kinase–substrate network (Fig. 2E), MAPK1, MAPK3, MAPK9, and RPS6KA3 were responsible for the phosphorylation of c-Fos at Ser374. Of these, the consensus phosphorylation motifs of MAPK3 and MAPK1, also known as extracellular signal–regulated protein kinases 1 and 2 (ERK1/2), matched the sequence pattern of c-Fos near Ser374, as identified via mass spectrometry (Fig. 4, A and B). This evidence suggests that ERK1/2 act as the serine/threonine kinases that phosphorylate c-Fos at Ser374. ERK1/2 play an important role in the mitogen-activated protein kinase (MAPK) signal transduction cascade, which is responsible for many biological functions (
). ERK1/2 are activated by various extracellular stimuli and are subsequently involved in cell proliferation, motility, and survival of cancer cells through the phosphorylation of their downstream transcriptional factors, including c-Fos (
). Therefore, we hypothesized that MSCs activate the ERK1/2–c-Fos signaling axis via the secretion of paracrine factors. The kinase activities of ERK1/2 depend on the phosphorylation status of the threonine 202 and tyrosine 204 residues (
). Therefore, we first validated whether MSC-CM could induce the phosphorylation of ERK1/2 at Thr202 and Tyr204. Western blotting showed that the phosphorylation levels of ERK1/2 at Thr202 and Tyr204 were upregulated in both LM cells (Fig. 4C) and A549 cells (Fig. 4D) after MSC-CM stimulation. These high phosphorylation ratios were normalized by the addition of glyceraldehyde 3-phosphate dehydrogenase (GAPDH), relative to the control group (Fig. 4, E and F). Combining these findings with our results described earlier, we speculated that ERK1/2 might play crucial roles in MSC-CM–mediated tumor progression by elevating the phosphorylation of c-Fos. Hence, SCH772984, as a specific ERK1/2 inhibitor, was applied with MSC-CM to lung cancer cells to suppress the kinase activities of ERK1/2 (
). As expected, SCH772984 cotreatment abrogated MSC-CM–induced cell proliferation (Fig. 4G) and cell migration (Fig. 4H) in the cells. Combining these results, we have illustrated that MSCs promote cell proliferation and motility in lung cancer through phosphorylation of the ERK1/2–c-Fos axis.
Fig. 4MSC-CM enhances tumor progression via the activation of the ERK/c-Fos signaling axis.A, the MS/MS spectrum of the phosphopeptide from c-Fos. The fragment ions in the MS2 spectrum localized at the phosphorylation site to serine 374. B, the consensus motifs of ERK1/2 (P-X-S/T-P) matched the sequence near the serine 374 site of c-Fos. C and D, the levels of ERK1/2 and phospho-ERK1/2-T202/Y204 were measured using Western blotting. Relative expression levels of phosphorylated ERK1/2 in cells stimulated with Control-CM or MSC-CM were normalized according to the level of total ERK1/2 and the loading control (GAPDH). E and F, the quantified results of Western blotting from three biological replicates are displayed as dot plots. G and H, cell proliferation (G) and cell migration (H) were detected in the lung cancer cells treated with or without MSC-CM in the presence or absence of ERK1/2 inhibitor (SCH772984). All quantified results are shown as mean ± standard deviation. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, two-tailed t test. CM, conditioned medium; ERK1/2, extracellular signal–regulated protein kinases 1 and 2; MSC, mesenchymalstem cell.
). We were curious about whether MSC-CM contains abundant eATP, which would regulate cancer development. We first attempted to elucidate whether ATP stimulation could lead to similar changes in the phosphorylation of the ERK1/2–c-Fos axis as were caused by MSC-CM. Western blotting indicated that the levels of phosphorylation of ERK1/2 at Thr202 and Tyr 204 and in c-Fos at Ser 374 were both increased after ATP stimulation (supplemental Fig. S1, A and B). The relative phosphorylated levels of phopho-ERK1/2-T202/Y204 and phospho-c-Fos-S374 in ATP-treated cells compared to the control group in each time point was normalized according to the level of total ERK1/2 or c-Fos and the loading control (GAPDH) (supplemental Fig. S1, C–F). Emerging evidence has proven that ATP synthase is ectopically expressed on the surface of various cell types (
). This cell-surface ATP synthase, also known as ecto-ATP synthase, contributes to the extracellular production of ATP in normal cells as well as cancer cells (
). To verify whether the accumulating eATP results from ecto-ATP synthase on the cell surface of MSCs, we performed an immunocytochemistry experiment using an antibody against ATP synthase in MSCs. With the cells in their nonpermeabilized form, only any ATP synthase on the cell surface would be detected, and intracellular ATP synthase would not. Cells that were permeabilized using TritonX-100 were used to establish a positive control for the antibody. In contract, cells incubated with the nonspecific antibody, IgG, did not show any green florescent signal. The results showed that the MSCs indeed expressed abundant ecto-ATP synthase, with a punctuated distribution, which was quite different from the distribution observed in mitochondrial networks (Fig. 5A). In contrast, the signal associated with intracellular ATP synthase in the permeabilized MSCs was almost entirely colocalized with the mitochondria. We then wished to know whether ecto-ATP synthase was commonly present on most MSCs. We therefore used flow cytometry to quantify the percentage of ecto-ATP synthase–positive MSCs. The data indicated that approximately 63% of MSCs expressed ecto-ATP synthase (Fig. 5B). An ATP bioluminescence assay also showed that the concentration of eATP in the MSC-CM was consistently higher than that in the CM of the cancer cells (LM-CM and A549-CM; Fig. 5C). This evidence implies that the higher eATP concentration in the TME might result from ecto-ATP synthase on the cell surface of MSCs. To further understand the influence of ecto-ATP synthase at the eATP level, we carried out citreoviridin, oligomycin, and carbonyl cyanide 3-chlorophenylhydrazone as the specific ecto-ATP synthase inhibitors for suppression of eATP release. In previous studies, these ecto-ATP synthase inhibitors are able to significantly decrease the concentration of eATP via the specific inhibitory effect on ecto-ATP synthase (
). As shown in Figure 5D, each ecto-ATP synthase inhibitor effectively reduced the concentration of eATP in MSCs, which was consistent with previous studies. Numerous studies point out that ectopic adenylate kinase (ecto-AK) also presents on the cell surface and contributes to the eATP level by catalyzing ATP + AMP from two ADP (
). In order to determine the effect of ecto-AK on eATP production, MSCs pretreated with ecto-ATP synthase inhibitors or a known ecto-AK inhibitor, Ap5A, were incubated with ADP alone. The eATP formation was only repressed by Ap5A after adding ADP (supplemental Fig. S2A). Although these data suggest the eATP level might partially be attributed to the activity of ecto-AK, another piece of evidence determined that the concentration of eATP level was significantly elevated in the presence of both ADP and Pi compared to ADP alone (supplemental Fig. S2B). These data illustrate that ecto-ATP synthase indeed has enzyme activity to promote eATP synthesis. Moreover, treating lung cancer cells with ecto-ATPi-MSC-CM, which contained less eATP, led to a slower proliferation rate than MSC-CM treatment (Fig. 5, E and F). Combining these results, we believe that ecto-ATP synthase on the cell surface of MSCs produces eATP into the TME and leads to the activation of the ERK1/2–c-Fos axis in lung cancer cells.
Fig. 5MSCs secrete extracellular ATP for activation of the ERK1/2–phospho-c-Fos pathway. A, the abundance of cell-surface and mitochondrial ATP synthase was assessed using immunocytochemistry with ATP-synthase–complex antibodies in nonpermeable and permeable MSCs, respectively. MitoTracker Red was used to label the mitochondria, and DAPI was used to stain the nuclei. Scale bar, 20 μm. B, ectopic ATP synthase–positive MSCs were assessed using flow cytometry with antibodies against the ATP synthase complex or control mouse IgG. In (A) and (B), MSCs incubated with mouse IgG antibodies served as the negative control, whereas MSCs permeabilized with 0.1% Triton X-100 as the positive control. C, the concentration of extracellular ATP in medium conditioned by lung cancer cells (LM and A549) and MSCs was measured and normalized according to the number of cells present. D, the extracellular ATP in the conditioned medium of MSCs treated with or without the ectopic ATP synthase inhibitors (citreoviridin, oligomycin, and CCCP, respectively) in the presence of ADP and Pi was detected. The relative eATP concentration was normalized according to the number of cells and the control group. E and F, cell proliferation was detected in the LM (E) and A549 (F) lung cancer cells treated with or without MSC-CM in the presence or absence of ecto-ATPi. All quantified results are shown as mean ± standard deviation. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, two-tailed t test. CCCP, carbonyl cyanide 3-chlorophenylhydrazone; CM, conditioned medium; LM, low-motility; MSC, mesenchymal stem cell.
eATP Secreted by MSCs Regulates Tumor Development via P2X7 Purinergic Receptor Activation
Growing evidence demonstrates that eATP in the TME activates P2 purinergic receptors that are commonly expressed on cancer cells. In the P2X receptor family, P2X purinergic receptor subtype 7 (P2X7R) acts as a central player in cancer development (
). The aforementioned result determined that ATP stimulation activated the phosphorylation of the ERK1/2–c-Fos axis in lung cancer cells (supplemental Fig. S1). We first attempted to uncover whether ATP-induced phosphorylation signaling further regulated proliferation and migration. As shown in supplemental Fig. S3, A and B, ATP treatment increased proliferation rates and motility abilities in LM and A549 lung cancer cells, whereas ATP and a P2X7R antagonist, A 438079, cotreatment reversed these effects. To further understand whether MSCs promote tumor progression by releasing eATP to stimulate P2X7Rs located on the plasma membrane of cancer cells, we used A 438079 to compete with eATP secreted by MSCs. Prior to stimulation with MSC-CM, lung cancer cells were pretreated with A 438079 in order to perform a P2X7R blockade. Western blotting showed that MSC-induced phosphorylation of the ERK1/2–c-Fos axis was abolished by the P2X7R blockade in both LM cells (Fig. 6, A–C) and A549 cells (Fig. 6, D–F). Moreover, the P2X7R blockade also suppressed the cell proliferation (Fig. 6, G and H) and cell migration (Fig. 6,I and J) that was enhanced by MSC-CM treatment. Taken together, these results suggest that eATP secreted by MSCs regulate tumor proliferation and motility via activation of the P2X7R–ERK1/2–c-Fos pathway.
Fig. 6Inhibition of the P2X7R on lung cancer cells abrogates MSC-CM-triggered tumor proliferation and migration.A–F, the levels of proteins and phosphoproteins of MSC-CM–treated lung cancer cells in the presence or absence of P2X7R inhibitor (A 438079). Relative expression levels of phosphoproteins were normalized according to the levels of total proteins and the loading control (GAPDH) and are shown as dot plots. G and H, the number of LM cells (G) and A549 cells (H) after treatment with or without MSC-CM in the presence or absence of A 438079 was counted at the indicated time points. I and J, cell migration was assessed in LM cells (I) and A549 cells (J) after treatment with or without MSC-CM in the presence or absence of A 438079. Representative images from three independent experiments are displayed, and the number of migrating cells is shown as mean ± standard deviation in histogram plots. ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001, two-tailed t test. CM, conditioned medium; LM, low-motility; MSC, mesenchymal stem cell.
During the last 10 years, the focus of cancer research has shifted away from the malignant tumor itself and to the TME and its interactions with the tumor. The TME is also thought to play a key role in tumor progression by recruiting stromal cells to provide growth factors and cytokines for tumor cell growth (
). MSCs are one of the stromal cell types in the TME, and emerging data indicate that it can boost tumor development and provide chemoresistance to anticancer drugs (
). In our study, we have demonstrated that MSC-CM improves the cell proliferation and migration abilities of lung cancer cells (Fig. 1). In addition, in order to elucidate the signaling pathways in lung cancer cells that are induced by MSCs, a quantitative phosphoproteomic profiling was performed to reveal MSC-triggered signaling pathway on the MSC-CM treated lung cancer cells (Fig. 2, A–E). The result showed that most of the more significantly expressed kinases are involved in the MAPK pathway, which mediates cell proliferation and cell motility (
). The MAPKs are classified into three families: extracellular signal–regulated kinase (ERK), Jun amino-terminal kinase (JNK), and p38/stress-activated protein kinase (p38/SAPK). The ERK pathway, also known as the p44/p42 MAPK pathway, is typically upregulated in human tumors and thus represents an attractive target for anticancer drug development (
). Our Western blotting analysis demonstrated that the expression of phospho-ERK1/2-T202/Y204 increased significantly in lung cancer cells stimulated with MSC-CM (Fig. 4, C–F). Moreover, the ERK1/2 inhibitor (SCH772984) abolished MSC-induced cell proliferation and migration in lung cancer cells (Fig. 4, G and H).
The activated ERKs translocate to the nucleus and phosphorylate transcription factors such as c-Fos, c-Jun, c-Myc, Elk-1, Sap-1a, Ets1, Tal, and CREB to regulate tumor progression (
Signal transduction mediated by the Ras/Raf/MEK/ERK pathway from cytokine receptors to transcription factors: Potential targeting for therapeutic intervention.
), acts as a member of the activated protein-1 (AP-1) family. It can be dimerized with c-Jun to form the transcription factor activator protein 1 (AP-1) complex, which can bind to DNA at AP-1-specific sites to enhance tumor growth (
). Herein, we first found that the phosphorylation level of c-Fos doubled in our quantitative phosphoproteomics investigations (Fig. 2E) and the consensus motif of the ERK1/2 substrate is similar to the sequence near c-Fos Ser374 (Fig. 4, A and B). Our Western blotting results further confirmed that MSCs triggered the MAPK pathway to phosphorylate c-Fos on serine 374 at various time points. We suggest that serine 374 on c-Fos plays a crucial role in MSC-mediated lung cancer cell function (Fig. 3, E–H).
A growing number of studies have reported that ecto-ATP synthase is overexpressed on the cell membrane of various cell types (
). In this study, we surprisingly observed that the whole ATP synthase complex that contains a full complement of subunits was expressed on the cell surface of MSCs using anti-ATP synthase complex immunocapture antibody. This ecto-ATP synthase with functional activity can generate eATP into the lung cancer microenvironment (Fig. 5). P2X7R, one of the receptors in purinergic signaling pathways, presents on the cell membranes of cancer cells, and its expression level is associated with tumor survival, proliferation, and metastasis (
Basal activation of the P2X7 ATP receptor elevates mitochondrial calcium and potential, increases cellular ATP levels, and promotes serum-independent growth.
P2X7 receptor antagonism inhibits p38 mitogen-activated protein kinase activation and ameliorates neuronal apoptosis after subarachnoid hemorrhage in rats.
). Consistent with these findings, our results indicate that ATP at a concentration of 0.5 mM increased the phosphorylation of c-Fos at serine 374 via activation of the MAPK pathway in lung cancer cells. Moreover, P2X7R blockade reduced MSC-induced c-Fos-S374 phosphorylation in the MAPK pathway of lung cancer cells and functionally downregulated MSC-induced proliferation and migration in lung cancer cells (Fig. 6). Therefore, an in-depth study of the purinergic signaling pathways of the TME may provide us with new cancer therapeutic targets.
Collectively, in this study, we aimed to dissect the role of MSCs in the microenvironment of lung cancer tumors. We discovered that ATP synthase was detected not only on the mitochondria but also on the cell surface of MSCs. As a consequence, ecto-ATP synthase was located on the cell membranes of MSCs and produced eATP into the lung cancer microenvironment. This eATP stimulates the P2X7R on lung cancer cells via ionotropic purinergic signaling. Finally, the eATP in MSC-CM triggered the MAPK pathway to phosphorylate c-Fos at its serine 374 residues to improve c-Fos stability in lung cancer cells. In conclusion, an MSC-derived TME promotes lung cancer cell progression via phosphorylation of c-Fos-S374 in the ERK/MAPK signaling axis (Fig. 7).
Fig. 7Schematic diagram of the phosphorylation mechanism induced by MSCs in lung cancer. Ectopic ATP synthase expressed on the cell surface of MSCs secretes extracellular ATP into the tumor microenvironment. This extracellular ATP activates P2X7R on lung cancer cells and subsequently induces the phospho-ERK1/2–phospho-c-Fos axis. The MSCs then contribute to lung cancer development via the promotion of tumor proliferation and migration. ERK1/2, extracellular signal–regulated protein kinases 1 and 2; MSC, mesenchymal stem cell.
Phospho-proteomic profile with MSC-CM stimulation in LM cells data have been deposited to ProteomeXchange (http://www.proteomexchange.org/) via the PRIDE partner repository with the dataset identifier PXD029288.
The authors declare that they have no conflict of interest.
Acknowledgments
We thank Technology Commons in College of Life Science, National Taiwan University, for technical assistance. We thank Prof. Jia-Lin Lee (Institute of Molecular and Cellular Biology, National Tsing Hua University) for providing LM cells used in this study and Prof. Chia-Lang Hsu (Department of Medical Research, National Taiwan University Hospital) for the constructive suggestions. The schematic figures were created with BioRender.com. We would also like to thank Uni-edit (www.uni-edit.net) and and Anthony Abram for editing and proofreading this manuscript.
Funding and additional information
This work was supported by the Ministry of Science and Technology (MOST 107–2314-B-002–254-MY3, MOST 109–2221-E-002–161-MY3, MOST 109–2320-B-002–017-MY3, MOST 109–2221-E-010–012-MY3, and MOST 109–2221-E-010–011-MY3), the Higher Education Sprout Project (NTU-109L8837A).
Author contributions
Y-. W. C., C-. C. W., C-. F. Y., and C-. H. W. investigation; Y-. W. C. formal analysis; Y-. W. C. and C-. F. Y. visualization; Y-. W. C., C-. C. W., and H-. F. J. writing-original draft; Y-. W. C., H-. C. H., and H-. F. J. writing-review and editing; H-. C. H. and H-. F. J. conceptualization; H-. C. H. and H-. F. J. funding acquisition; H-. C. H. and H-. F. J. project administration.
Mesenchymal stem cells isolated from human gliomas increase proliferation and maintain stemness of glioma stem cells through the IL-6/gp130/STAT3 pathway.
Basal activation of the P2X7 ATP receptor elevates mitochondrial calcium and potential, increases cellular ATP levels, and promotes serum-independent growth.
P2X7 receptor antagonism inhibits p38 mitogen-activated protein kinase activation and ameliorates neuronal apoptosis after subarachnoid hemorrhage in rats.
Establishment of immortalized mesenchymal stromal cells with red fluorescence protein expression for in vivo transplantation and tracing in the rat model with traumatic brain injury.
A functionally active human F1F0 ATPase can be purified by immunocapture from heart tissue and fibroblast cell lines. Subunit structure and activity studies.
Signal transduction mediated by the Ras/Raf/MEK/ERK pathway from cytokine receptors to transcription factors: Potential targeting for therapeutic intervention.
00045-7&id=fx1.jpg){kind=link}
00045-7&id=gr1.jpg){kind=link}
00045-7&id=gr2.jpg){kind=link}
00045-7&id=gr3.jpg){kind=link}
00045-7&id=gr4.jpg){kind=link}
00045-7&id=gr5.jpg){kind=link}
00045-7&id=gr6.jpg){kind=link}
00045-7&id=gr7.jpg){kind=link}