Molecular BioSystems View Article Online
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
PAPER
Cite this: Mol. BioSyst., 2014, 10, 1146
View Journal | View Issue
Characterization of secreted vesicles from vascular smooth muscle cells† Laura Comelli,a Silvia Rocchiccioli,*a Salvatore Smirni,a Alessandra Salvetti,b Giovanni Signore,c Lorenzo Citti,a Maria Giovanna Trivellaa and Antonella Cecchettiniab The artery medial layer is mainly composed of vascular smooth muscle cells (VSMCs). These cells contribute to the formation of neointima and atherosclerotic plaques by switching from the quiescent-contractile to migratory-activated state. Apoptotic blebs, microvesicles and exosomes are secreted vesicles, with differences in composition and size, involved in cellular communication at multiple levels. In this article, an untargeted, proteomics approach was exploited to characterise VSMC released vesicles and a preliminary protein profile for microvesicles and exosomes of different cell phenotypes was obtained. Enriched samples of vesicles from serum-free and serum-activated VSMCs were analysed by a LC-MS/MS strategy leading to the identification of 349 proteins. In microvesicles, the most abundant classes of identified proteins were cytoplasmic or organelle associated, house keeping and metabolic factors. Otherwise, exosomes from different phenotypes revealed a sharper peculiarity thus, as suggested by the high percentage of ECM and ECM related proteins and cell adhesion molecules, they seem to play an important role in outward or cell-to-cell signalling. Comparison
Received 30th November 2013, Accepted 26th February 2014
between exosomes or microvesicles from quiescent and activated VSMCs evidenced 29 differentially expressed
DOI: 10.1039/c3mb70544g
proteins. Among these, in microvesicles there are several proteins that are involved in vesicle trafficking while in exosomes focal adhesion and ECM related factors are the most interesting. These data, although
www.rsc.org/molecularbiosystems
preliminary, are promising for a possible identification of potential circulating markers of a cell state.
Introduction The artery medial layer is mainly composed of circumferentially aligned vascular smooth muscle cells (VSMCs) interspersed within elastic fibres. These cells are characterised by a great plasticity that enables them to respond to several stimuli by switching from their quiescent contractile phenotype to a more migratory, proliferative, synthetic and activated state.1 In response to environmental signals, VSMCs activate important molecular and morphological changes such as altered expression of contractile proteins and extracellular matrix components, synthesis of cell markers, proteases and inflammatory cytokines.2 VSMC primary cultures represent a useful and validated cell model since they switch from a synthetic, activated phenotype to a quiescent, contractile one merely by serum withdrawal.3 Vascular injury is a triggering event for a
Institute of Clinical Physiology, CNR, Pisa, Italy. E-mail: [email protected] b Unit of Experimental Biology and Genetics, Department of Clinical and Experimental Medicine, University of Pisa, Pisa, Italy c Center for Nanotechnology Innovation @ NEST, Istituto Italiano di Tecnologia, Pisa, Italy † Electronic supplementary information (ESI) available. See DOI: 10.1039/ c3mb70544g
1146 | Mol. BioSyst., 2014, 10, 1146--1152
the acquisition of an activated i.e. pathological phenotype by VSMCs that, as a consequence, deeply contributes to the onset of vascular proliferative disorders.4 Due to the substantial involvement of VSMCs in cardiovascular pathology, both as structural and functional elements, special attention has been focused on the analysis of modulated factors during cell activation.5 VSMCs are also responsible for autocrine/paracrine signals towards other cells, and since soluble proteins are in charge for cell communication, the study of secreted proteins is an important field of research. Particularly it is of interest the vesicle mediated intercellular communication system through which bio-active molecules such as RNAs, microRNAs and proteins may be transferred to target cells.6 Based on differences in morphology and formation mechanisms, cell released vesicles have been classified into apoptotic blebs, microvesicles and exosomes.7,8 Apoptotic blebs, as indicated by the name, are released during the process of programmed cell death, thus they represent fragments of dying cells ranging from 50 to 500 nm in diameter. Microvesicles are heterogeneous, greater than 100 nm diameter vesicles, which shed from the plasma membrane, while exosomes are smaller structures, ranging from 30 to 100 nm in diameter, and originate from intracellular multivesicular bodies.9 Over the last few years it has been extensively demonstrated that all these vesicles are secreted either in vitro by cultured cells,
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
Paper
or in vivo by organs and tissues. The presence of circulating exosomes and microvesicles in many biological fluids, including blood, urine, saliva, milk, suggests their participation in intercellular communication and in the modulation of different physiological processes.10 For these reasons, the study of secreted vesicles represents a growing field of interest from a diagnostic point of view, aiming at identifying disease biomarkers in biological fluids. Secreted proteins include extracellular matrix components as well as proteins shedding from the cell surface or the cell cytoplasm and reflecting the instantaneous phenotype of cells at various stages of development, differentiation or disease progression. It is strongly accepted that they represent an important source of biomarkers for pathophysiological events.11 The proteomics study of the whole mixture of compounds that is released by cells, tissues or organs (comprehensive of soluble proteins, apoptotic blebs, microvesicles and exosomes) is referred to as secretomics.12 Its analysis is complicated by the presence of serum-derived, high abundant proteins that impair or reduce the detection of relatively low represented proteins. In this article, VSMC released vesicles were characterised by a gel-free untargeted proteomics approach. Apoptotic blebs, microvesicles and exosomes from both serum-free cultured (thus quiescent) and serumactivated VSMCs were analysed and a total of 349 proteins were identified. The comparison of proteins between microvesicles and exosomes showed a microvesicle up-regulation of factors responsible for cell housekeeping, metabolic and secretory pathways. On the other hand, exosome involvement in cell communication has been confirmed by the high percentage of ECM and ECM related proteins and cell adhesion molecules identified. It is noteworthy that the results presented here, although preliminary, also suggest a differential secretion of some proteins in quiescent cells with respect to activated cells, thus they are promising for the identification of potential circulating markers of a cell state.
Materials and methods Isolation of VSMCs and culture conditions Medial VSMCs were isolated by dissection of coronaries from the myocardium of 8-month-old domestic crossbred pigs (Sus scrofa domesticus). Enzymatic digestion and cultivation were performed according to the method described by Christen et al.13 The experiments reported were done with VSMCs inside the sixth passage and, unless otherwise specified, from at least three different explants. VSMCs were cultured under standard conditions in DMEM medium supplemented with 10% FBS (activated-VSMCs) and left in serum-free medium for 48 h (quiescent-VSMCs). For vesicle isolation, approximately 106 cells were seeded into 100 mm diameter culture dishes (14 dishes per experiment) containing 7 mL of culture medium and cultured until cell density reached 90% confluence. 100 000 g ultracentrifuged FBS was used to avoid bovine vesicle contaminations. Vesicle isolation Conditioned medium was collected and centrifuged at 300 g to eliminate cell debris. The pellet was discarded and for secretome
This journal is © The Royal Society of Chemistry 2014
Molecular BioSystems
preparation the sample was stored at 80 1C until analysis. Samples were concentrated by centrifugal devices Amicon Ultra-3 (Millipore) following the manufacturer’s recommendations. Protein concentration was determined by bicinchoninic acid (Pierce). In vesicle separation the supernatant was centrifuged at 2000 g for 30 minutes (min) to obtain apoptotic blebs and at 10 000 g for 1 h at 4 1C to obtain microvesicles. The supernatant was subject to filtration using 0.22 mm filters to eliminate debris and ultracentrifuged at 100 000 g for 2 h at 4 1C to obtain exosomes. Each pellet was washed with PBS. Dynamic light scattering analysis The three fractions of vesicles (apoptotic blebs, microvesicles and exosomes) from serum-free quiescent VSMCs were diluted to 25 mg mL 1 to obtain an optimum scattering intensity. Measurements were performed on a Malvern zetasizer nano (Malvern Instruments Ltd, UK) employing a 633 nm laser. Samples were analysed at 25 1C and scattering was detected at 901 to the incident beam. Twenty records were acquired for each fraction and data were analysed by Malvern software to obtain vesicle diameter. Transmission electron microscopy Fractions were placed on 200-mesh formvar carbon coated copper grids. The sample was stained with five drops of 2.5% (w/v) of uranyl acetate solution and incubated for 2 min at room temperature. The excess solution was removed by blotting the edge of each grid onto a filter paper and the grid was air dried for 30 min. Samples were examined using a Jeol 100 SX transmission electron microscope. Protein extraction and tryptic digestion The vesicle pellet was frozen at 196 1C in liquid nitrogen for 30 s and then thawed at 50 1C for 2 min. The freeze–thaw processes were repeated for 5 cycles and then samples were sonicated 5 min. The protein concentration was determined by bicinchoninic acid (BCA) and 40 mg of each sample were processed. Reduction was performed with a final concentration of 5 mM DTT at 80 1C for 20 min. Samples were allowed to cool at room temperature and alkylation was done for 30 min at 37 1C after addition of iodoacetamide up to 10 mM final concentration. Digestion was performed adding 0.50 mg of trypsin at 37 1C overnight. Western blot analysis Extracts (20 mg) were run on a 10% SDS-PAGE, the proteins were transferred onto a nitrocellulose membrane (Amersham) using a wet transfer system (Biorad). The following antibodies were used: anti-TSG101 mouse monoclonal antibody (1 : 500) (Santa Cruz Biotechnology, Inc), anti-Flotillin-1 mouse monoclonal antibody (1 : 1000) (Santa Cruz Biotechnology, Inc), anti-CD81 goat polyclonal antibody (1 : 500) (Santa Cruz Biotechnology, Inc), anti-Prohibitin mouse monoclonal antibody (1 : 200) (Santa Cruz Biotechnology, Inc), anti-Vinculin goat polyclonal antibody (1 : 500) (Santa Cruz Biotechnology, Inc), anti-HSP70L1 rabbit monoclonal antibody (1 : 500) (Epitomics).
Mol. BioSyst., 2014, 10, 1146--1152 | 1147
View Article Online
Molecular BioSystems
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
LC-MS/MS analysis 40 mg of the sample were analyzed using a TripleTOFt 5600 mass spectrometer (AB SCIEX, Toronto, Canada) interfaced with an Ultimate 3000 nano-HPLC system (LC Packings, DIONEX, USA). 50 mL of filtrate samples were added to 50 mL of a solution composed of 2% ACN and 0.1% formic acid. The loading pump pre-concentrated the sample in a pre-column cartridge (PepMap-100 C18 5 mm 100 A, 30 mm id 5 mm). Chromatographic separation of peptides was performed using a C18 PepMap-100 column (15 cm 75 mm id, LC Packings DIONEX) equilibrated at 45 1C with a solvent A (water/acetonitrile 98/2 vol/vol, 0.1% formic acid) at a flow rate of 300 nL min 1. Runs were performed under a 60 min linear gradient from 10 to 45% of solvent B (water/acetonitrile 2/98 vol/vol, 0.1% formic acid) followed by a 10 min purge step at 95% of B before a 20 min re-equilibration step to the starting conditions. The column was directly coupled to a TripleTOFt 5600 System (AB SCIEX, Toronto, Canada), equipped with a DuoSprayt ion source (AB SCIEX, Toronto, Canada). The mass spectrometer was controlled by Analysts 1.6.1 software (AB SCIEX, Toronto, Canada). Data were acquired using an ion spray voltage of 3 kV, curtain gas set at 25, GS1 10 and GS2 0 PSI nitrogen flow, source temperature 150 1C. For information dependent acquisition (IDA) analysis, survey scans were acquired in 250 ms and 25 product ion scans were collected if exceeding a threshold of 125 counts per second (counts per s). Four time bins were summed for each scan at a pulser frequency value of 11 kHz through monitoring of the 40 GHz multichannel TDC detector with four-anode/channel detection. Dynamic exclusion was set for 1/2 of peak width (B8 s), and then the precursor was refreshed off the exclusion list.
Paper
from serum-activated VSMCs were collected from 90% confluent cultures. Activated cells were cultured in medium supplemented with FBS that had been previously depleted from any endogenous vesicles. Cell-released vesicles were obtained by differential centrifugations and preparative ultracentrifugations as described in Material and methods. Different fractions were analysed by transmission electron microscopy and the relative images in Fig. 1A show vesicles of expected dimensions that are representative of apoptotic bodies, microvesicles and exosomes. In order to better dimensionally characterise the vesicles, also dynamic light scattering analyses were performed on the three samples. The size distribution across twenty measurements for each vesicle population was 423 99 nm (Mean SD) for apoptotic blebs enriched fraction, 181 52 nm for microvesicles and 28 6 nm for exosomes (Fig. 1B). The enrichment protocol, adopted for the isolation of the three different vesicle populations, was assessed by Western blots, probing the membranes with specific antibodies for well-known markers. TSG-101, Flotillin-1 and CD81, which are
Protein identification and data analysis MS/MS data were processed with ProteinPilott software (AB SCIEX, Toronto, Canada), using the Paragont and Pro Groupt Algorithms and SwissProt 2012 as protein databases for Sus scrofa specie. The false discovery rate (FDR) analysis was done using the integrated tools in ProteinPilot software and a confidence level of 95% was set. The statistical comparative analysis was performed using MarkerViewt software 1.2.1 (AB SCIEX). The ion chromatograms of high confidence peptides identified by ProteinPilott were extracted using PeakViewt software and then MS peak areas and identifications were imported into MarkerViewt software. Normalization of the total protein content was obtained using a global normalization of profiles using MarkerViewt 1.2 software. Principal Component Analysis (PCA) was performed on mass spectrometric data in order to evidence groupings among the data set. The groups were compared with the t-test using a threshold of 95% ( p value o 0.05) and log fold change 40.5 and o 0.5.
Results Isolation and characterisation of VSMC-derived exosomes, microvesicles and apoptotic blebs Apoptotic blebs, microvesicles and exosomes from quiescent VSMCs, i.e. cells cultured for 48 h in serum-free medium, and
1148 | Mol. BioSyst., 2014, 10, 1146--1152
Fig. 1 Characterization of VSMC-derived vesicles. (A) Representative transmission electron micrographs showing isolated apoptotic blebs, microvesicles and exosomes. (B) DLS analysis of apoptotic blebs, microvesicles and exosomes. Media values and SD are reported. Gaussian curves are generated by Origin 8.0 software using averaged values from twenty independent measurements. (C) Representative Western blots of secretome, apoptotic blebs, microvesicles, exosomes and total cell extract lysates (20 mg per well).
This journal is © The Royal Society of Chemistry 2014
View Article Online
Paper
Molecular BioSystems
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
widely accepted as markers for microvesicles and exosomes, were detected in the expected fractions, while they resulted absent in apoptotic blebs (Fig. 1C), supporting the workflow suitability. Also the complete secretome, i.e. the global released protein mixture including soluble as well as vesicle-associated proteins, was analysed and a weaker signal, probably due to the vesicle dilution, was observed in the corresponding lanes. Finally and as further support, Prohibitin, which is a mitochondrial marker, was not detected in the exosome fraction. Proteomics analysis of released vesicles Proteomes of isolated vesicles from quiescent and activated VSMCs were characterized by LC-MS/MS using a 5600 Triple-TOF mass spectrometer and analysed by Protein Pilott software (ABSCIEX) with a FDR less than 1%. A total of 349 proteins were identified of which 171 proteins were identified in apoptotic blebs, 127 proteins in microvesicles, 51 proteins in exosomes (Tables 1S–6S, ESI†). LC runs were performed in triplicate and all samples were grouped on the basis of their common features according to the PCA that was performed by MarkerViewt software (ABSCIEX). Interestingly, protein data from apoptotic blebs of both quiescent and activated cells clustered in close proximity. In contrast, proteins from microvesicles and exosomes resulted to be more differentiated. Using Uniprot database annotations, microvesicle and exosome identified proteins were classified according to their cellular localisation and grouped as follows: cytoplasmic, mitochondrial, secreted, membrane, endoplasmic reticulum, Golgi apparatus, nuclear, extracellular matrix (Fig. 2A). On the basis of their functions, proteins were grouped as follows: cell adhesion, extracellular matrix (ECM), cell signalling, metabolic pathways, chaperones, cytoskeleton organization, vesicle trafficking and housekeeping proteins (Fig. 2B). In microvesicles, most of the proteins are cytoplasmic or membrane associated and, with regard to the functional classification, a considerable fraction is represented by housekeeping factors, involved in the metabolic pathway and in cell signalling. Otherwise, in the exosomal fractions, as far as cell localisation is concerned, even if cytoplasmic components are still present, a noticeable enrichment of ECM and secreted proteins can be observed, while the membrane components are considerably reduced. The functional classification also evidenced an increase in proteins involved in ECM, as well as in the cell adhesion and secretory pathway. In order to better characterise the identified proteins, we exploited the SecretomeP algorithm. This tool, checking for the presence of signal fragments in the peptide sequences, discriminates between classically (i.e. through endoplasmic reticulum) and non-classically secreted proteins. From this analysis, the majority (57–62%) of the identified proteins, either in microvesicles or in exosomes, resulted predicted as secreted, with an enrichment of the non-classically secreted fraction in the microvesicles. Since a percentage ranging between 38% and 44% resulted non predicted as secreted and in order to validate our findings, we compared our results with the data published in ExoCarta,14 a database repository listing proteins extracted from exosome-related research publications. This analysis revealed that many of the identified proteins had been reported in previous studies on exosomes, namely 67% in exosomes and 47% of proteins in microvesicles. In brief, our identifications
This journal is © The Royal Society of Chemistry 2014
Fig. 2 Classification of identified microvesicle and exosome proteins from quiescent and activated VSMCs. (A) Protein cellular localization; cytoplasmic (Cyt), mitochondrial (Mito), secreted (Sec), membrane (Membr), endoplasmic reticulum (ER), Golgi apparatus (Golgi), nuclear (Nucl), extracellular matrix (ECM). (B) Protein biological functions; cytoskeleton organization (Cyt org), chaperones (CH), cell adhesion (CA), cell signalling (CS), metabolic pathways (MP), vesicle trafficking (Vtraf), extracellular matrix and related (ECM&rel) and housekeeping proteins (HK). (C) Representative Western blots of differentially modulated proteins.
seem to be congruent with data originated from other exosome studies and thus the vesicle-mediated secretory origin of the identified proteins is assessed. Microvesicle- and exosome-derived proteins from quiescent and activated VSMCs Statistical analyses of the acquired data were performed by MarkerViewt software that, by the alignment of mass and
Mol. BioSyst., 2014, 10, 1146--1152 | 1149
View Article Online
Molecular BioSystems
Paper
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
Table 1 Differentially represented microvesicle proteins from quiescent vs. activated VSMCs. Protein name and Accession number were reported according to the SwissProt 2012 database. Cell function was referred to data reported in recent literature. FC stays for Fold Change values and the minus sign indicates down-regulation of the protein; .: down regulation and m: up regulation in microvesicles from quiescent VSMCs when compared with microvesicles from activated VSMCs. p-Values were generated using MarkerView software. NC Sec: Non Classical Secretion; C Sec: Classical Secretion; NP: Non Predicted secretion, according to SecretomeP
Protein name
Accession number
Cell function
Quiescent vs. activated FC
Annexin A1 Flotillin-1 Rab GDP dissociation inhibitor beta Ras-related protein Rab-1B Protein S100-A10 Annexin A2 Leukocyte surface antigen CD47 Guanine nucleotide-binding protein G(s) subunit alpha Tubulin alpha-1B Tubulin alpha-1A Glyceraldehyde-3-phosphate dehydrogenase Aminopeptidase N D3phosphoglycerate dehydrogenase Superoxide dismutase Malate dehydrogenase
P19619 Q767L6 Q6Q7J2 Q06AU7 P04163 P19620 Q9GKE8 P29797 Q2XVP4 P02550 P00355 P15145 A5GFY8 P04178 P00346
Membrane fusion; exocytosis Caveolae formation Vesicle formation Vesicle formation Vesicle formation Exocytosis Cell adhesion Cell signalling Cytoskeleton organization Cytoskeleton organization Metabolic pathway Metalloprotease Metabolic pathway Antioxidant Metabolic pathway
4.4 4.5 9.8 18.2 26.4 4.3 14.2 10.2 8.1 7.2 6.9 4.2 7.3 3.8 9.8
m m m m m m m m m m m m m m m
p-Value
SecretomeP
0.03563 0.0069 0.01024 0.04909 0.04728 0.01244 0.00502 0.0473 0.00317 0.03687 0.03714 0.00086 0.00168 0.03739 0.00518
NC Sec NP NP NP NP NC Sec NC Sec NC Sec NP NP NC Sec C Sec NP NC Sec NC Sec
Table 2 Differentially represented exosome proteins from quiescent vs. activated VSMCs. Protein name and Accession number were reported according to the SwissProt 2012 database. Cell function was referred to data reported in recent literature. FC stays for Fold Change values and the minus sign indicates downregulation of the protein; .: down regulation and m: up regulation in exosomes from quiescent VSMCs compared with exosomes from activated VSMCs. p-Values were generated using MarkerView software. NC Sec: Non Classical Secretion; C Sec: Classical Secretion; NP: Non Predicted secretion, according to SecretomeP
Protein name
Accession number
Cell function
Biglycan Fibrillin-1 Hyaluronan and proteoglycan link protein 1 Secreted phosphoprotein 24 Tenascin Integrin beta-1 Vinculin Inter-alpha-trypsin inhibitor heavy chain H2 Signal transducer and activator of transcription 3 Peptidyl-prolyl cis–trans isomerase A Heat shock 70 kDa protein 1-like Glyceraldehyde-3-phosphate dehydrogenase 40S ribosomal protein S23 Cofilin-2
Q9GKQ6 Q9TV36 P10859 Q711S8 Q29116 Q9GLP0 P26234 O02668 Q19S50 P62936 A5A8V7 P00355 Q6SA96 Q5G6V9
ECM component ECM component ECM component ECM component ECM and Focal adhesion Focal adhesion Focal adhesion Focal adhesion and cell growth Growth signaling pathway Protein folding Protein folding Metabolic pathway House keeping Cytoskeleton organization
retention time of spectral peaks, allows determination of up- and down-regulated components in the samples. Comparative analyses between proteins identified in microvesicles and exosomes from quiescent vs. activated VSMCs evidenced the differential expression ( p o 0.05, log fold of change 4 0.5 and o 0.5) of 15 proteins in microvesicles (Table 1) and 14 in exosomes (Table 2). This analysis was performed on technical triplicates for each sample. Identified proteins were further divided into up- and down-regulated (quiescent vs. activated cells) according to their log fold of change 40.5 and o 0.5 respectively. To validate the differential expression of the proteins determined by mass spectrometry analysis, Western blot assay was applied for Flotillin-1, HSP70 and Vinculin (Fig. 2C).
Discussion Secreted factors play a crucial role in cell-to-cell communication, which, coordinating important biological functions such
1150 | Mol. BioSyst., 2014, 10, 1146--1152
Quiescent vs. activated FC 9.2 m 4.4 m 5.7 . 24.4 . 5.2 m 5.2 m 3.9 m 6.5 m 27 . 9.5 . 6.9 . 15.9 . 7.1 . 13 .
p-Value
SecretomeP
0.00058 0.01542 0.0099 0.01102 0.02202 0.00323 0.03273 0.01564 0.03776 0.00806 0.0001 0.03251 0.03882 0.00224
C Sec C Sec C Sec C Sec C Sec NP NP C Sec NC Sec NP NP NC Sec C Sec NC Sec
as growth, differentiation, apoptosis and signalling cascade regulation, is the key regulator of tissue homeostasis. Apoptotic blebs, microvesicles and exosomes represent a sub-fraction of the secretome and it has been clearly shown, mostly in cancer biology, that they are involved in multiple levels of intercellular communication. As expected, the apoptotic bodies include DNA, RNA, and histones, and display signal molecules that help a rapid clearance.15 In contrast, microvesicles and exosomes appeared to be more specific in their composition and displayed differences related to the type of vesicle and to the originating phenotype. Additionally, microvesicle and exosome analyses revealed a large amount of the identified proteins already present in the ExoCarta database,14 data that support our separation method, and also highlight the identification of a relevant percentage of new microvesicular and exosomal proteins. The classification based on Uniprot database annotations of the identified proteins offers interesting cues for reflection. In microvesicles, the proteins more represented are cytoplasmic
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
Paper
Molecular BioSystems
or organelle associated (Golgi, ER, mitochondria) and functionally house keeping enzymes and factors. This is not surprising due to the unselective, budding mechanism of formation. Conversely, in exosomes, secreted or ECM related proteins and focal adhesion molecules are the most representative groups. Of note, no mitochondrial proteins were identified in exosomes, while the nuclear proteins are more numerous than in microvesicles. This last observation is not negligible and we do not consider these data as artefacts since nuclear proteins could have important roles in transcription regulation of exosome target cells. When we focus on the comparison between vesicles released from quiescent or activated VSMCs we observe a completely different protein profiling and it comes out that proteins related to cytoskeleton organization and the metabolic pathway are in percentage more represented in both microvesicles and exosomes of activated cells, while housekeeping and cell signalling proteins are more represented in quiescent vesicles. Table 3 briefly summarises these observations, schematically recounting the most represented classes of proteins in exosomes and microvesicles from quiescent and activated VSMCs. It is evident that exosomes from different phenotypes reveal a sharper peculiarity and, as suggested by the high percentage of ECM and ECM related proteins and cell adhesion molecules, they seem to play an important role in outwards or cell-to-cell signalling. The differences between vesicles from quiescent and activated VSMCs suggested by the protein profiling were checked by a comparative study, and at first glance the PCA revealed a similar protein composition in apoptotic blebs from both VSMC phenotypes, in accordance with the assumption that these vesicles were released as a consequence of a common mechanism during the late stages of cell death, irrespective of the previous cell phenotype, and derived from apoptotic cell fragmentation into subcellular apoptotic bodies. On the other hand, it is noteworthy the documentation of a differential molecular composition in microvesicles and exosomes from quiescent and activated cells, suggesting the possibility of identifying potential circulating markers of a cell state. Expressly, some of the identified proteins that are involved in vesicular trafficking (Flotillin-1, Ras related Rab-1B, Rab GDP dissociation inhibitor beta, Annexin A1, Annexin A2, Protein S100-A10) increased in microvesicles from quiescent VSMCs with respect to microvesicles from activated cells. A particular comment is necessary for the proteins that are modulated in exosomes from quiescent with respect to activated VSMCs. The significant over-representation of focal adhesion proteins and ECM proteins (Inter-alpha-trypsin inhibitor, Tenascin,
Table 3 Schematic summary of the most abundant classes of identified proteins. The most represented classes of proteins in exosomes and microvesicles from quiescent and activated VSMCs are indicated
Quiescent Activated
Microvesicles
Exosomes
Cytoplasm/organelles House keeping Vesicle trafficking Membrane Cytoskeleton organization Metabolic pathways
ECM and related Cell adhesion
This journal is © The Royal Society of Chemistry 2014
Nucleus Chaperones
Integrin, Vinculin, Byglican, Fibrillin-1) suggests that the interaction among contractile cells, being accomplished through the focal adhesion proteins and ECM components, may be mediated by the exosomes. Indeed, these results confirmed what already published about the over-regulation of focal adhesion proteins in quiescent, contractile vs. activated VSMCs.16 Conversely, exosomes from activated VSMCs display another set of up-regulated proteins mainly devoted to stress signalling, cell growth regulation and cell mobility (Stat-3, Hsp-70, Peptidyl-prolyl cis–trans isomerase A, Glyceraldehyde-3-phosphate dehydrogenase, Cofilin-2). These observations suggested a stronger role of exosomes, with respect to microvesicles, in functional signalling during VSMC phenotype change. The role of VSMC phenotype switch in the onset and development of cardiovascular diseases is widely accepted and demonstrated.17 Therefore, it is noteworthy that exosomes from activated VSMCs display a potentially active/ activable Stat-3-dependent stress signalling pathway including a putative upstream role of PPIA in JAK/STAT phosphorylation18 and the downstream involvement of HSP70. In brief, our results suggest the inclusion in exosomes of a set of stress signals that once delivered into target cells or tissues might extensively contribute to vascular remodelling.
Conclusions A fractionation protocol enabled us to obtain enriched fractions that are representative of different vesicles, as demonstrated not only on the basis of displayed markers and morphological properties but also in light of completely different protein profiles. It is conceivable that properties of the two kinds of vesicles (microvesicles and exosomes) may produce considerably different environmental signals. Microvesicles, budding from cell membranes, seem to reflect mostly an instantaneous gathering of elements in the surroundings of VSMC plasma membranes, including cell cortex content (factors that are localised in proximity of the cell surface, i.e. directly underneath the cell membrane) and membrane-associated proteins involved in vesicular formation and trafficking. On the other hand, exosomes, which are produced by more specific mechanisms, seem to recruit defined panels of proteins whose assortment would depend on the phenotypical status of the cells. To the best of our knowledge, this is the first proteomics study performed on secreted vesicles from quiescent and activated VSMCs. Given that a lot of proteins that were identified in cell derived microvesicles and exosomes can have a role in VSMC migration and proliferation, their characterisation could provide a better understanding of the intercellular communication signals leading to the phenotype switch. In this respect, exosomes look more intriguing than microvesicles displaying sets of proteins involved in more strategic cell-tocell signalling. This is a preliminary analysis and more accurate studies are necessary in order to set a panel of characteristics able to assess utilization of microvesicles and exosomes as potential biomarkers of physiological or pathological phenotype conditions.
Mol. BioSyst., 2014, 10, 1146--1152 | 1151
View Article Online
Molecular BioSystems
Acknowledgements This work was partially funded by the ARTreat-FP7 project (Grant Agreement FP7 224297) and by the Fondazione Cassa di Risparmio di Pisa in the framework of the Micro-VAST project.
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
References 1 J. B. Michel, Z. Li and P. Lacolley, Smooth muscle cells and vascular diseases, Cardiovasc. Res., 2012, 95, 135–137. 2 A. Tedgui and Z. Mallat, Cytokines in atherosclerosis: pathogenic and regulatory pathways, Physiol. Rev., 2006, 86, 515–581. 3 L. Poliseno, A. Cecchettini, L. Mariani, M. Evangelista, F. Ricci, F. Giorgi, L. Citti and G. Rainaldi, Resting smooth muscle cells as a model for studying vascular cell activation, Tissue Cell, 2006, 38, 111–120. 4 A. Cecchettini, S. Rocchiccioli, C. Boccardi and L. Citti, Vascular smooth-muscle-cell activation: proteomics point of view, Int. Rev. Cell Mol. Biol., 2011, 288, 43–99. 5 Y. Nakagawa, K. Ikeda, Y. Akakabe, M. Koide, M. Uraoka, K. T. Yutaka, R. Kurimoto-Nakano, T. Takahashi, S. Matoba and H. Yamada, et al. Paracrine osteogenic signals via bone morphogenetic protein-2 accelerate the atherosclerotic intimal calcification in vivo, Arterioscler., Thromb., Vasc. Biol., 2010, 30, 1908–1915. 6 E. Hergenreider, S. Heydt, K. Treguer, T. Boettger, A. J. Horrevoets, A. M. Zeiher, M. P. Scheffer, A. S. Frangakis, X. Yin and M. Mayr, et al. Atheroprotective communication between endothelial cells and smooth muscle cells through miRNAs, Nat. Cell Biol., 2012, 14, 249–256. 7 S. Mathivanan, H. Ji and R. J. Simpson, Exosomes: extracellular organelles important in intercellular communication, J. Proteomics, 2010, 73, 1907–1920. 8 C. Thery, M. Ostrowski and E. Segura, Membrane vesicles as conveyors of immune responses, Nat. Rev. Immunol., 2009, 9, 581–593.
1152 | Mol. BioSyst., 2014, 10, 1146--1152
Paper
9 V. Ramakrishnaiah, C. Thumann, I. Fofana, F. Habersetzer, Q. Pan, P. E. de Ruiter, R. Willemsen, J. A. Demmers, V. Stalin Raj and G. Jenster, et al. Exosome-mediated transmission of hepatitis C virus between human hepatoma Huh7.5 cells, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 13109–13113. 10 A. Lakkaraju and E. Rodriguez-Boulan, Itinerant exosomes: emerging roles in cell and tissue polarity, Trends Cell Biol., 2008, 18, 199–209. 11 F. Raimondo, L. Morosi, C. Chinello, F. Magni and M. Pitto, Advances in membranous vesicle and exosome proteomics improving biological understanding and biomarker discovery, Proteomics, 2011, 11, 709–720. 12 Y. Hathout, Approaches to the study of the cell secretome, Expert Rev. Proteomics, 2007, 4, 239–248. 13 T. Christen, M. L. Bochaton-Piallat, P. Neuville, S. Rensen, M. Redard, G. van Eys and G. Gabbiani, Circ. Res., 1999, 85, 99–107. 14 S. Mathivanan and R. J. Simpson, ExoCarta: A compendium of exosomal proteins and RNA, Proteomics, 2009, 9, 4997–5000. 15 G. Wickman, L. Julian and M. F. Olson, How apoptotic cells aid in the removal of their own cold dead bodies, Cell Death Differ., 2012, 19, 735–742. 16 S. Rocchiccioli, N. Ucciferri, L. Comelli, M. G. Trivella, L. Citti and A. Cecchettini, Proteomics changes in adhesion molecules: a driving force for vascular smooth muscle cell phenotypic switch, Mol. BioSyst., 2012, 8, 1052–1059. 17 S. Rocchiccioli, G. Pelosi, S. Rosini, M. Marconi, F. Viglione, L. Citti, M. Ferrari, M. G. Trivella and A. Cecchettini, Secreted proteins from carotid endarterectomy: an untargeted approach to disclose molecular clues of plaque progression, J. Transl. Med., 2013, 11, 260. 18 Z. Tian-Tian, Z. Jun-Feng and G. Heng, Functions of cyclophilin A in atherosclerosis, Exp. Clin. Cardiol., 2013, 18, e118–e124.
This journal is © The Royal Society of Chemistry 2014
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
PAPER
Cite this: Mol. BioSyst., 2014, 10, 1146
View Journal | View Issue
Characterization of secreted vesicles from vascular smooth muscle cells† Laura Comelli,a Silvia Rocchiccioli,*a Salvatore Smirni,a Alessandra Salvetti,b Giovanni Signore,c Lorenzo Citti,a Maria Giovanna Trivellaa and Antonella Cecchettiniab The artery medial layer is mainly composed of vascular smooth muscle cells (VSMCs). These cells contribute to the formation of neointima and atherosclerotic plaques by switching from the quiescent-contractile to migratory-activated state. Apoptotic blebs, microvesicles and exosomes are secreted vesicles, with differences in composition and size, involved in cellular communication at multiple levels. In this article, an untargeted, proteomics approach was exploited to characterise VSMC released vesicles and a preliminary protein profile for microvesicles and exosomes of different cell phenotypes was obtained. Enriched samples of vesicles from serum-free and serum-activated VSMCs were analysed by a LC-MS/MS strategy leading to the identification of 349 proteins. In microvesicles, the most abundant classes of identified proteins were cytoplasmic or organelle associated, house keeping and metabolic factors. Otherwise, exosomes from different phenotypes revealed a sharper peculiarity thus, as suggested by the high percentage of ECM and ECM related proteins and cell adhesion molecules, they seem to play an important role in outward or cell-to-cell signalling. Comparison
Received 30th November 2013, Accepted 26th February 2014
between exosomes or microvesicles from quiescent and activated VSMCs evidenced 29 differentially expressed
DOI: 10.1039/c3mb70544g
proteins. Among these, in microvesicles there are several proteins that are involved in vesicle trafficking while in exosomes focal adhesion and ECM related factors are the most interesting. These data, although
www.rsc.org/molecularbiosystems
preliminary, are promising for a possible identification of potential circulating markers of a cell state.
Introduction The artery medial layer is mainly composed of circumferentially aligned vascular smooth muscle cells (VSMCs) interspersed within elastic fibres. These cells are characterised by a great plasticity that enables them to respond to several stimuli by switching from their quiescent contractile phenotype to a more migratory, proliferative, synthetic and activated state.1 In response to environmental signals, VSMCs activate important molecular and morphological changes such as altered expression of contractile proteins and extracellular matrix components, synthesis of cell markers, proteases and inflammatory cytokines.2 VSMC primary cultures represent a useful and validated cell model since they switch from a synthetic, activated phenotype to a quiescent, contractile one merely by serum withdrawal.3 Vascular injury is a triggering event for a
Institute of Clinical Physiology, CNR, Pisa, Italy. E-mail: [email protected] b Unit of Experimental Biology and Genetics, Department of Clinical and Experimental Medicine, University of Pisa, Pisa, Italy c Center for Nanotechnology Innovation @ NEST, Istituto Italiano di Tecnologia, Pisa, Italy † Electronic supplementary information (ESI) available. See DOI: 10.1039/ c3mb70544g
1146 | Mol. BioSyst., 2014, 10, 1146--1152
the acquisition of an activated i.e. pathological phenotype by VSMCs that, as a consequence, deeply contributes to the onset of vascular proliferative disorders.4 Due to the substantial involvement of VSMCs in cardiovascular pathology, both as structural and functional elements, special attention has been focused on the analysis of modulated factors during cell activation.5 VSMCs are also responsible for autocrine/paracrine signals towards other cells, and since soluble proteins are in charge for cell communication, the study of secreted proteins is an important field of research. Particularly it is of interest the vesicle mediated intercellular communication system through which bio-active molecules such as RNAs, microRNAs and proteins may be transferred to target cells.6 Based on differences in morphology and formation mechanisms, cell released vesicles have been classified into apoptotic blebs, microvesicles and exosomes.7,8 Apoptotic blebs, as indicated by the name, are released during the process of programmed cell death, thus they represent fragments of dying cells ranging from 50 to 500 nm in diameter. Microvesicles are heterogeneous, greater than 100 nm diameter vesicles, which shed from the plasma membrane, while exosomes are smaller structures, ranging from 30 to 100 nm in diameter, and originate from intracellular multivesicular bodies.9 Over the last few years it has been extensively demonstrated that all these vesicles are secreted either in vitro by cultured cells,
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
Paper
or in vivo by organs and tissues. The presence of circulating exosomes and microvesicles in many biological fluids, including blood, urine, saliva, milk, suggests their participation in intercellular communication and in the modulation of different physiological processes.10 For these reasons, the study of secreted vesicles represents a growing field of interest from a diagnostic point of view, aiming at identifying disease biomarkers in biological fluids. Secreted proteins include extracellular matrix components as well as proteins shedding from the cell surface or the cell cytoplasm and reflecting the instantaneous phenotype of cells at various stages of development, differentiation or disease progression. It is strongly accepted that they represent an important source of biomarkers for pathophysiological events.11 The proteomics study of the whole mixture of compounds that is released by cells, tissues or organs (comprehensive of soluble proteins, apoptotic blebs, microvesicles and exosomes) is referred to as secretomics.12 Its analysis is complicated by the presence of serum-derived, high abundant proteins that impair or reduce the detection of relatively low represented proteins. In this article, VSMC released vesicles were characterised by a gel-free untargeted proteomics approach. Apoptotic blebs, microvesicles and exosomes from both serum-free cultured (thus quiescent) and serumactivated VSMCs were analysed and a total of 349 proteins were identified. The comparison of proteins between microvesicles and exosomes showed a microvesicle up-regulation of factors responsible for cell housekeeping, metabolic and secretory pathways. On the other hand, exosome involvement in cell communication has been confirmed by the high percentage of ECM and ECM related proteins and cell adhesion molecules identified. It is noteworthy that the results presented here, although preliminary, also suggest a differential secretion of some proteins in quiescent cells with respect to activated cells, thus they are promising for the identification of potential circulating markers of a cell state.
Materials and methods Isolation of VSMCs and culture conditions Medial VSMCs were isolated by dissection of coronaries from the myocardium of 8-month-old domestic crossbred pigs (Sus scrofa domesticus). Enzymatic digestion and cultivation were performed according to the method described by Christen et al.13 The experiments reported were done with VSMCs inside the sixth passage and, unless otherwise specified, from at least three different explants. VSMCs were cultured under standard conditions in DMEM medium supplemented with 10% FBS (activated-VSMCs) and left in serum-free medium for 48 h (quiescent-VSMCs). For vesicle isolation, approximately 106 cells were seeded into 100 mm diameter culture dishes (14 dishes per experiment) containing 7 mL of culture medium and cultured until cell density reached 90% confluence. 100 000 g ultracentrifuged FBS was used to avoid bovine vesicle contaminations. Vesicle isolation Conditioned medium was collected and centrifuged at 300 g to eliminate cell debris. The pellet was discarded and for secretome
This journal is © The Royal Society of Chemistry 2014
Molecular BioSystems
preparation the sample was stored at 80 1C until analysis. Samples were concentrated by centrifugal devices Amicon Ultra-3 (Millipore) following the manufacturer’s recommendations. Protein concentration was determined by bicinchoninic acid (Pierce). In vesicle separation the supernatant was centrifuged at 2000 g for 30 minutes (min) to obtain apoptotic blebs and at 10 000 g for 1 h at 4 1C to obtain microvesicles. The supernatant was subject to filtration using 0.22 mm filters to eliminate debris and ultracentrifuged at 100 000 g for 2 h at 4 1C to obtain exosomes. Each pellet was washed with PBS. Dynamic light scattering analysis The three fractions of vesicles (apoptotic blebs, microvesicles and exosomes) from serum-free quiescent VSMCs were diluted to 25 mg mL 1 to obtain an optimum scattering intensity. Measurements were performed on a Malvern zetasizer nano (Malvern Instruments Ltd, UK) employing a 633 nm laser. Samples were analysed at 25 1C and scattering was detected at 901 to the incident beam. Twenty records were acquired for each fraction and data were analysed by Malvern software to obtain vesicle diameter. Transmission electron microscopy Fractions were placed on 200-mesh formvar carbon coated copper grids. The sample was stained with five drops of 2.5% (w/v) of uranyl acetate solution and incubated for 2 min at room temperature. The excess solution was removed by blotting the edge of each grid onto a filter paper and the grid was air dried for 30 min. Samples were examined using a Jeol 100 SX transmission electron microscope. Protein extraction and tryptic digestion The vesicle pellet was frozen at 196 1C in liquid nitrogen for 30 s and then thawed at 50 1C for 2 min. The freeze–thaw processes were repeated for 5 cycles and then samples were sonicated 5 min. The protein concentration was determined by bicinchoninic acid (BCA) and 40 mg of each sample were processed. Reduction was performed with a final concentration of 5 mM DTT at 80 1C for 20 min. Samples were allowed to cool at room temperature and alkylation was done for 30 min at 37 1C after addition of iodoacetamide up to 10 mM final concentration. Digestion was performed adding 0.50 mg of trypsin at 37 1C overnight. Western blot analysis Extracts (20 mg) were run on a 10% SDS-PAGE, the proteins were transferred onto a nitrocellulose membrane (Amersham) using a wet transfer system (Biorad). The following antibodies were used: anti-TSG101 mouse monoclonal antibody (1 : 500) (Santa Cruz Biotechnology, Inc), anti-Flotillin-1 mouse monoclonal antibody (1 : 1000) (Santa Cruz Biotechnology, Inc), anti-CD81 goat polyclonal antibody (1 : 500) (Santa Cruz Biotechnology, Inc), anti-Prohibitin mouse monoclonal antibody (1 : 200) (Santa Cruz Biotechnology, Inc), anti-Vinculin goat polyclonal antibody (1 : 500) (Santa Cruz Biotechnology, Inc), anti-HSP70L1 rabbit monoclonal antibody (1 : 500) (Epitomics).
Mol. BioSyst., 2014, 10, 1146--1152 | 1147
View Article Online
Molecular BioSystems
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
LC-MS/MS analysis 40 mg of the sample were analyzed using a TripleTOFt 5600 mass spectrometer (AB SCIEX, Toronto, Canada) interfaced with an Ultimate 3000 nano-HPLC system (LC Packings, DIONEX, USA). 50 mL of filtrate samples were added to 50 mL of a solution composed of 2% ACN and 0.1% formic acid. The loading pump pre-concentrated the sample in a pre-column cartridge (PepMap-100 C18 5 mm 100 A, 30 mm id 5 mm). Chromatographic separation of peptides was performed using a C18 PepMap-100 column (15 cm 75 mm id, LC Packings DIONEX) equilibrated at 45 1C with a solvent A (water/acetonitrile 98/2 vol/vol, 0.1% formic acid) at a flow rate of 300 nL min 1. Runs were performed under a 60 min linear gradient from 10 to 45% of solvent B (water/acetonitrile 2/98 vol/vol, 0.1% formic acid) followed by a 10 min purge step at 95% of B before a 20 min re-equilibration step to the starting conditions. The column was directly coupled to a TripleTOFt 5600 System (AB SCIEX, Toronto, Canada), equipped with a DuoSprayt ion source (AB SCIEX, Toronto, Canada). The mass spectrometer was controlled by Analysts 1.6.1 software (AB SCIEX, Toronto, Canada). Data were acquired using an ion spray voltage of 3 kV, curtain gas set at 25, GS1 10 and GS2 0 PSI nitrogen flow, source temperature 150 1C. For information dependent acquisition (IDA) analysis, survey scans were acquired in 250 ms and 25 product ion scans were collected if exceeding a threshold of 125 counts per second (counts per s). Four time bins were summed for each scan at a pulser frequency value of 11 kHz through monitoring of the 40 GHz multichannel TDC detector with four-anode/channel detection. Dynamic exclusion was set for 1/2 of peak width (B8 s), and then the precursor was refreshed off the exclusion list.
Paper
from serum-activated VSMCs were collected from 90% confluent cultures. Activated cells were cultured in medium supplemented with FBS that had been previously depleted from any endogenous vesicles. Cell-released vesicles were obtained by differential centrifugations and preparative ultracentrifugations as described in Material and methods. Different fractions were analysed by transmission electron microscopy and the relative images in Fig. 1A show vesicles of expected dimensions that are representative of apoptotic bodies, microvesicles and exosomes. In order to better dimensionally characterise the vesicles, also dynamic light scattering analyses were performed on the three samples. The size distribution across twenty measurements for each vesicle population was 423 99 nm (Mean SD) for apoptotic blebs enriched fraction, 181 52 nm for microvesicles and 28 6 nm for exosomes (Fig. 1B). The enrichment protocol, adopted for the isolation of the three different vesicle populations, was assessed by Western blots, probing the membranes with specific antibodies for well-known markers. TSG-101, Flotillin-1 and CD81, which are
Protein identification and data analysis MS/MS data were processed with ProteinPilott software (AB SCIEX, Toronto, Canada), using the Paragont and Pro Groupt Algorithms and SwissProt 2012 as protein databases for Sus scrofa specie. The false discovery rate (FDR) analysis was done using the integrated tools in ProteinPilot software and a confidence level of 95% was set. The statistical comparative analysis was performed using MarkerViewt software 1.2.1 (AB SCIEX). The ion chromatograms of high confidence peptides identified by ProteinPilott were extracted using PeakViewt software and then MS peak areas and identifications were imported into MarkerViewt software. Normalization of the total protein content was obtained using a global normalization of profiles using MarkerViewt 1.2 software. Principal Component Analysis (PCA) was performed on mass spectrometric data in order to evidence groupings among the data set. The groups were compared with the t-test using a threshold of 95% ( p value o 0.05) and log fold change 40.5 and o 0.5.
Results Isolation and characterisation of VSMC-derived exosomes, microvesicles and apoptotic blebs Apoptotic blebs, microvesicles and exosomes from quiescent VSMCs, i.e. cells cultured for 48 h in serum-free medium, and
1148 | Mol. BioSyst., 2014, 10, 1146--1152
Fig. 1 Characterization of VSMC-derived vesicles. (A) Representative transmission electron micrographs showing isolated apoptotic blebs, microvesicles and exosomes. (B) DLS analysis of apoptotic blebs, microvesicles and exosomes. Media values and SD are reported. Gaussian curves are generated by Origin 8.0 software using averaged values from twenty independent measurements. (C) Representative Western blots of secretome, apoptotic blebs, microvesicles, exosomes and total cell extract lysates (20 mg per well).
This journal is © The Royal Society of Chemistry 2014
View Article Online
Paper
Molecular BioSystems
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
widely accepted as markers for microvesicles and exosomes, were detected in the expected fractions, while they resulted absent in apoptotic blebs (Fig. 1C), supporting the workflow suitability. Also the complete secretome, i.e. the global released protein mixture including soluble as well as vesicle-associated proteins, was analysed and a weaker signal, probably due to the vesicle dilution, was observed in the corresponding lanes. Finally and as further support, Prohibitin, which is a mitochondrial marker, was not detected in the exosome fraction. Proteomics analysis of released vesicles Proteomes of isolated vesicles from quiescent and activated VSMCs were characterized by LC-MS/MS using a 5600 Triple-TOF mass spectrometer and analysed by Protein Pilott software (ABSCIEX) with a FDR less than 1%. A total of 349 proteins were identified of which 171 proteins were identified in apoptotic blebs, 127 proteins in microvesicles, 51 proteins in exosomes (Tables 1S–6S, ESI†). LC runs were performed in triplicate and all samples were grouped on the basis of their common features according to the PCA that was performed by MarkerViewt software (ABSCIEX). Interestingly, protein data from apoptotic blebs of both quiescent and activated cells clustered in close proximity. In contrast, proteins from microvesicles and exosomes resulted to be more differentiated. Using Uniprot database annotations, microvesicle and exosome identified proteins were classified according to their cellular localisation and grouped as follows: cytoplasmic, mitochondrial, secreted, membrane, endoplasmic reticulum, Golgi apparatus, nuclear, extracellular matrix (Fig. 2A). On the basis of their functions, proteins were grouped as follows: cell adhesion, extracellular matrix (ECM), cell signalling, metabolic pathways, chaperones, cytoskeleton organization, vesicle trafficking and housekeeping proteins (Fig. 2B). In microvesicles, most of the proteins are cytoplasmic or membrane associated and, with regard to the functional classification, a considerable fraction is represented by housekeeping factors, involved in the metabolic pathway and in cell signalling. Otherwise, in the exosomal fractions, as far as cell localisation is concerned, even if cytoplasmic components are still present, a noticeable enrichment of ECM and secreted proteins can be observed, while the membrane components are considerably reduced. The functional classification also evidenced an increase in proteins involved in ECM, as well as in the cell adhesion and secretory pathway. In order to better characterise the identified proteins, we exploited the SecretomeP algorithm. This tool, checking for the presence of signal fragments in the peptide sequences, discriminates between classically (i.e. through endoplasmic reticulum) and non-classically secreted proteins. From this analysis, the majority (57–62%) of the identified proteins, either in microvesicles or in exosomes, resulted predicted as secreted, with an enrichment of the non-classically secreted fraction in the microvesicles. Since a percentage ranging between 38% and 44% resulted non predicted as secreted and in order to validate our findings, we compared our results with the data published in ExoCarta,14 a database repository listing proteins extracted from exosome-related research publications. This analysis revealed that many of the identified proteins had been reported in previous studies on exosomes, namely 67% in exosomes and 47% of proteins in microvesicles. In brief, our identifications
This journal is © The Royal Society of Chemistry 2014
Fig. 2 Classification of identified microvesicle and exosome proteins from quiescent and activated VSMCs. (A) Protein cellular localization; cytoplasmic (Cyt), mitochondrial (Mito), secreted (Sec), membrane (Membr), endoplasmic reticulum (ER), Golgi apparatus (Golgi), nuclear (Nucl), extracellular matrix (ECM). (B) Protein biological functions; cytoskeleton organization (Cyt org), chaperones (CH), cell adhesion (CA), cell signalling (CS), metabolic pathways (MP), vesicle trafficking (Vtraf), extracellular matrix and related (ECM&rel) and housekeeping proteins (HK). (C) Representative Western blots of differentially modulated proteins.
seem to be congruent with data originated from other exosome studies and thus the vesicle-mediated secretory origin of the identified proteins is assessed. Microvesicle- and exosome-derived proteins from quiescent and activated VSMCs Statistical analyses of the acquired data were performed by MarkerViewt software that, by the alignment of mass and
Mol. BioSyst., 2014, 10, 1146--1152 | 1149
View Article Online
Molecular BioSystems
Paper
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
Table 1 Differentially represented microvesicle proteins from quiescent vs. activated VSMCs. Protein name and Accession number were reported according to the SwissProt 2012 database. Cell function was referred to data reported in recent literature. FC stays for Fold Change values and the minus sign indicates down-regulation of the protein; .: down regulation and m: up regulation in microvesicles from quiescent VSMCs when compared with microvesicles from activated VSMCs. p-Values were generated using MarkerView software. NC Sec: Non Classical Secretion; C Sec: Classical Secretion; NP: Non Predicted secretion, according to SecretomeP
Protein name
Accession number
Cell function
Quiescent vs. activated FC
Annexin A1 Flotillin-1 Rab GDP dissociation inhibitor beta Ras-related protein Rab-1B Protein S100-A10 Annexin A2 Leukocyte surface antigen CD47 Guanine nucleotide-binding protein G(s) subunit alpha Tubulin alpha-1B Tubulin alpha-1A Glyceraldehyde-3-phosphate dehydrogenase Aminopeptidase N D3phosphoglycerate dehydrogenase Superoxide dismutase Malate dehydrogenase
P19619 Q767L6 Q6Q7J2 Q06AU7 P04163 P19620 Q9GKE8 P29797 Q2XVP4 P02550 P00355 P15145 A5GFY8 P04178 P00346
Membrane fusion; exocytosis Caveolae formation Vesicle formation Vesicle formation Vesicle formation Exocytosis Cell adhesion Cell signalling Cytoskeleton organization Cytoskeleton organization Metabolic pathway Metalloprotease Metabolic pathway Antioxidant Metabolic pathway
4.4 4.5 9.8 18.2 26.4 4.3 14.2 10.2 8.1 7.2 6.9 4.2 7.3 3.8 9.8
m m m m m m m m m m m m m m m
p-Value
SecretomeP
0.03563 0.0069 0.01024 0.04909 0.04728 0.01244 0.00502 0.0473 0.00317 0.03687 0.03714 0.00086 0.00168 0.03739 0.00518
NC Sec NP NP NP NP NC Sec NC Sec NC Sec NP NP NC Sec C Sec NP NC Sec NC Sec
Table 2 Differentially represented exosome proteins from quiescent vs. activated VSMCs. Protein name and Accession number were reported according to the SwissProt 2012 database. Cell function was referred to data reported in recent literature. FC stays for Fold Change values and the minus sign indicates downregulation of the protein; .: down regulation and m: up regulation in exosomes from quiescent VSMCs compared with exosomes from activated VSMCs. p-Values were generated using MarkerView software. NC Sec: Non Classical Secretion; C Sec: Classical Secretion; NP: Non Predicted secretion, according to SecretomeP
Protein name
Accession number
Cell function
Biglycan Fibrillin-1 Hyaluronan and proteoglycan link protein 1 Secreted phosphoprotein 24 Tenascin Integrin beta-1 Vinculin Inter-alpha-trypsin inhibitor heavy chain H2 Signal transducer and activator of transcription 3 Peptidyl-prolyl cis–trans isomerase A Heat shock 70 kDa protein 1-like Glyceraldehyde-3-phosphate dehydrogenase 40S ribosomal protein S23 Cofilin-2
Q9GKQ6 Q9TV36 P10859 Q711S8 Q29116 Q9GLP0 P26234 O02668 Q19S50 P62936 A5A8V7 P00355 Q6SA96 Q5G6V9
ECM component ECM component ECM component ECM component ECM and Focal adhesion Focal adhesion Focal adhesion Focal adhesion and cell growth Growth signaling pathway Protein folding Protein folding Metabolic pathway House keeping Cytoskeleton organization
retention time of spectral peaks, allows determination of up- and down-regulated components in the samples. Comparative analyses between proteins identified in microvesicles and exosomes from quiescent vs. activated VSMCs evidenced the differential expression ( p o 0.05, log fold of change 4 0.5 and o 0.5) of 15 proteins in microvesicles (Table 1) and 14 in exosomes (Table 2). This analysis was performed on technical triplicates for each sample. Identified proteins were further divided into up- and down-regulated (quiescent vs. activated cells) according to their log fold of change 40.5 and o 0.5 respectively. To validate the differential expression of the proteins determined by mass spectrometry analysis, Western blot assay was applied for Flotillin-1, HSP70 and Vinculin (Fig. 2C).
Discussion Secreted factors play a crucial role in cell-to-cell communication, which, coordinating important biological functions such
1150 | Mol. BioSyst., 2014, 10, 1146--1152
Quiescent vs. activated FC 9.2 m 4.4 m 5.7 . 24.4 . 5.2 m 5.2 m 3.9 m 6.5 m 27 . 9.5 . 6.9 . 15.9 . 7.1 . 13 .
p-Value
SecretomeP
0.00058 0.01542 0.0099 0.01102 0.02202 0.00323 0.03273 0.01564 0.03776 0.00806 0.0001 0.03251 0.03882 0.00224
C Sec C Sec C Sec C Sec C Sec NP NP C Sec NC Sec NP NP NC Sec C Sec NC Sec
as growth, differentiation, apoptosis and signalling cascade regulation, is the key regulator of tissue homeostasis. Apoptotic blebs, microvesicles and exosomes represent a sub-fraction of the secretome and it has been clearly shown, mostly in cancer biology, that they are involved in multiple levels of intercellular communication. As expected, the apoptotic bodies include DNA, RNA, and histones, and display signal molecules that help a rapid clearance.15 In contrast, microvesicles and exosomes appeared to be more specific in their composition and displayed differences related to the type of vesicle and to the originating phenotype. Additionally, microvesicle and exosome analyses revealed a large amount of the identified proteins already present in the ExoCarta database,14 data that support our separation method, and also highlight the identification of a relevant percentage of new microvesicular and exosomal proteins. The classification based on Uniprot database annotations of the identified proteins offers interesting cues for reflection. In microvesicles, the proteins more represented are cytoplasmic
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
Paper
Molecular BioSystems
or organelle associated (Golgi, ER, mitochondria) and functionally house keeping enzymes and factors. This is not surprising due to the unselective, budding mechanism of formation. Conversely, in exosomes, secreted or ECM related proteins and focal adhesion molecules are the most representative groups. Of note, no mitochondrial proteins were identified in exosomes, while the nuclear proteins are more numerous than in microvesicles. This last observation is not negligible and we do not consider these data as artefacts since nuclear proteins could have important roles in transcription regulation of exosome target cells. When we focus on the comparison between vesicles released from quiescent or activated VSMCs we observe a completely different protein profiling and it comes out that proteins related to cytoskeleton organization and the metabolic pathway are in percentage more represented in both microvesicles and exosomes of activated cells, while housekeeping and cell signalling proteins are more represented in quiescent vesicles. Table 3 briefly summarises these observations, schematically recounting the most represented classes of proteins in exosomes and microvesicles from quiescent and activated VSMCs. It is evident that exosomes from different phenotypes reveal a sharper peculiarity and, as suggested by the high percentage of ECM and ECM related proteins and cell adhesion molecules, they seem to play an important role in outwards or cell-to-cell signalling. The differences between vesicles from quiescent and activated VSMCs suggested by the protein profiling were checked by a comparative study, and at first glance the PCA revealed a similar protein composition in apoptotic blebs from both VSMC phenotypes, in accordance with the assumption that these vesicles were released as a consequence of a common mechanism during the late stages of cell death, irrespective of the previous cell phenotype, and derived from apoptotic cell fragmentation into subcellular apoptotic bodies. On the other hand, it is noteworthy the documentation of a differential molecular composition in microvesicles and exosomes from quiescent and activated cells, suggesting the possibility of identifying potential circulating markers of a cell state. Expressly, some of the identified proteins that are involved in vesicular trafficking (Flotillin-1, Ras related Rab-1B, Rab GDP dissociation inhibitor beta, Annexin A1, Annexin A2, Protein S100-A10) increased in microvesicles from quiescent VSMCs with respect to microvesicles from activated cells. A particular comment is necessary for the proteins that are modulated in exosomes from quiescent with respect to activated VSMCs. The significant over-representation of focal adhesion proteins and ECM proteins (Inter-alpha-trypsin inhibitor, Tenascin,
Table 3 Schematic summary of the most abundant classes of identified proteins. The most represented classes of proteins in exosomes and microvesicles from quiescent and activated VSMCs are indicated
Quiescent Activated
Microvesicles
Exosomes
Cytoplasm/organelles House keeping Vesicle trafficking Membrane Cytoskeleton organization Metabolic pathways
ECM and related Cell adhesion
This journal is © The Royal Society of Chemistry 2014
Nucleus Chaperones
Integrin, Vinculin, Byglican, Fibrillin-1) suggests that the interaction among contractile cells, being accomplished through the focal adhesion proteins and ECM components, may be mediated by the exosomes. Indeed, these results confirmed what already published about the over-regulation of focal adhesion proteins in quiescent, contractile vs. activated VSMCs.16 Conversely, exosomes from activated VSMCs display another set of up-regulated proteins mainly devoted to stress signalling, cell growth regulation and cell mobility (Stat-3, Hsp-70, Peptidyl-prolyl cis–trans isomerase A, Glyceraldehyde-3-phosphate dehydrogenase, Cofilin-2). These observations suggested a stronger role of exosomes, with respect to microvesicles, in functional signalling during VSMC phenotype change. The role of VSMC phenotype switch in the onset and development of cardiovascular diseases is widely accepted and demonstrated.17 Therefore, it is noteworthy that exosomes from activated VSMCs display a potentially active/ activable Stat-3-dependent stress signalling pathway including a putative upstream role of PPIA in JAK/STAT phosphorylation18 and the downstream involvement of HSP70. In brief, our results suggest the inclusion in exosomes of a set of stress signals that once delivered into target cells or tissues might extensively contribute to vascular remodelling.
Conclusions A fractionation protocol enabled us to obtain enriched fractions that are representative of different vesicles, as demonstrated not only on the basis of displayed markers and morphological properties but also in light of completely different protein profiles. It is conceivable that properties of the two kinds of vesicles (microvesicles and exosomes) may produce considerably different environmental signals. Microvesicles, budding from cell membranes, seem to reflect mostly an instantaneous gathering of elements in the surroundings of VSMC plasma membranes, including cell cortex content (factors that are localised in proximity of the cell surface, i.e. directly underneath the cell membrane) and membrane-associated proteins involved in vesicular formation and trafficking. On the other hand, exosomes, which are produced by more specific mechanisms, seem to recruit defined panels of proteins whose assortment would depend on the phenotypical status of the cells. To the best of our knowledge, this is the first proteomics study performed on secreted vesicles from quiescent and activated VSMCs. Given that a lot of proteins that were identified in cell derived microvesicles and exosomes can have a role in VSMC migration and proliferation, their characterisation could provide a better understanding of the intercellular communication signals leading to the phenotype switch. In this respect, exosomes look more intriguing than microvesicles displaying sets of proteins involved in more strategic cell-tocell signalling. This is a preliminary analysis and more accurate studies are necessary in order to set a panel of characteristics able to assess utilization of microvesicles and exosomes as potential biomarkers of physiological or pathological phenotype conditions.
Mol. BioSyst., 2014, 10, 1146--1152 | 1151
View Article Online
Molecular BioSystems
Acknowledgements This work was partially funded by the ARTreat-FP7 project (Grant Agreement FP7 224297) and by the Fondazione Cassa di Risparmio di Pisa in the framework of the Micro-VAST project.
Published on 26 February 2014. Downloaded by University of Windsor on 27/10/2014 23:00:18.
References 1 J. B. Michel, Z. Li and P. Lacolley, Smooth muscle cells and vascular diseases, Cardiovasc. Res., 2012, 95, 135–137. 2 A. Tedgui and Z. Mallat, Cytokines in atherosclerosis: pathogenic and regulatory pathways, Physiol. Rev., 2006, 86, 515–581. 3 L. Poliseno, A. Cecchettini, L. Mariani, M. Evangelista, F. Ricci, F. Giorgi, L. Citti and G. Rainaldi, Resting smooth muscle cells as a model for studying vascular cell activation, Tissue Cell, 2006, 38, 111–120. 4 A. Cecchettini, S. Rocchiccioli, C. Boccardi and L. Citti, Vascular smooth-muscle-cell activation: proteomics point of view, Int. Rev. Cell Mol. Biol., 2011, 288, 43–99. 5 Y. Nakagawa, K. Ikeda, Y. Akakabe, M. Koide, M. Uraoka, K. T. Yutaka, R. Kurimoto-Nakano, T. Takahashi, S. Matoba and H. Yamada, et al. Paracrine osteogenic signals via bone morphogenetic protein-2 accelerate the atherosclerotic intimal calcification in vivo, Arterioscler., Thromb., Vasc. Biol., 2010, 30, 1908–1915. 6 E. Hergenreider, S. Heydt, K. Treguer, T. Boettger, A. J. Horrevoets, A. M. Zeiher, M. P. Scheffer, A. S. Frangakis, X. Yin and M. Mayr, et al. Atheroprotective communication between endothelial cells and smooth muscle cells through miRNAs, Nat. Cell Biol., 2012, 14, 249–256. 7 S. Mathivanan, H. Ji and R. J. Simpson, Exosomes: extracellular organelles important in intercellular communication, J. Proteomics, 2010, 73, 1907–1920. 8 C. Thery, M. Ostrowski and E. Segura, Membrane vesicles as conveyors of immune responses, Nat. Rev. Immunol., 2009, 9, 581–593.
1152 | Mol. BioSyst., 2014, 10, 1146--1152
Paper
9 V. Ramakrishnaiah, C. Thumann, I. Fofana, F. Habersetzer, Q. Pan, P. E. de Ruiter, R. Willemsen, J. A. Demmers, V. Stalin Raj and G. Jenster, et al. Exosome-mediated transmission of hepatitis C virus between human hepatoma Huh7.5 cells, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 13109–13113. 10 A. Lakkaraju and E. Rodriguez-Boulan, Itinerant exosomes: emerging roles in cell and tissue polarity, Trends Cell Biol., 2008, 18, 199–209. 11 F. Raimondo, L. Morosi, C. Chinello, F. Magni and M. Pitto, Advances in membranous vesicle and exosome proteomics improving biological understanding and biomarker discovery, Proteomics, 2011, 11, 709–720. 12 Y. Hathout, Approaches to the study of the cell secretome, Expert Rev. Proteomics, 2007, 4, 239–248. 13 T. Christen, M. L. Bochaton-Piallat, P. Neuville, S. Rensen, M. Redard, G. van Eys and G. Gabbiani, Circ. Res., 1999, 85, 99–107. 14 S. Mathivanan and R. J. Simpson, ExoCarta: A compendium of exosomal proteins and RNA, Proteomics, 2009, 9, 4997–5000. 15 G. Wickman, L. Julian and M. F. Olson, How apoptotic cells aid in the removal of their own cold dead bodies, Cell Death Differ., 2012, 19, 735–742. 16 S. Rocchiccioli, N. Ucciferri, L. Comelli, M. G. Trivella, L. Citti and A. Cecchettini, Proteomics changes in adhesion molecules: a driving force for vascular smooth muscle cell phenotypic switch, Mol. BioSyst., 2012, 8, 1052–1059. 17 S. Rocchiccioli, G. Pelosi, S. Rosini, M. Marconi, F. Viglione, L. Citti, M. Ferrari, M. G. Trivella and A. Cecchettini, Secreted proteins from carotid endarterectomy: an untargeted approach to disclose molecular clues of plaque progression, J. Transl. Med., 2013, 11, 260. 18 Z. Tian-Tian, Z. Jun-Feng and G. Heng, Functions of cyclophilin A in atherosclerosis, Exp. Clin. Cardiol., 2013, 18, e118–e124.
This journal is © The Royal Society of Chemistry 2014
