Quantitative Proteomics of Extracellular Vesicles Released from

Mar 26, 2014 - Vesicular secretion remains one of the most important paths for the proteins to exit the cell. Here, we have used high-throughput quant...
0 downloads 0 Views 4MB Size
Article pubs.acs.org/jpr

Quantitative Proteomics of Extracellular Vesicles Released from Human Monocyte-Derived Macrophages upon β‑Glucan Stimulation Wojciech Cypryk,† Tiina Ö hman,† Eeva-Liisa Eskelinen,‡ Sampsa Matikainen,§ and Tuula A. Nyman*,† †

Institute of Biotechnology, University of Helsinki, 00100 Helsinki, Finland Department of Biosciences, University of Helsinki, 00100 Helsinki, Finland § Finnish Institute of Occupational Health, 00250 Helsinki, Finland ‡

S Supporting Information *

ABSTRACT: Fungal infections (mycoses) are common diseases of varying severity that cause problems, especially to immunologically compromised people. Fungi express a variety of pathogen-associated molecular patterns on their surface including β-glucans, which are important immunostimulatory components of fungal cell walls. During stimulatory conditions of infection and colonization, besides intensive intracellular response, human cells actively communicate on the intercellular level by secreting proteins and other biomolecules with several mechanisms. Vesicular secretion remains one of the most important paths for the proteins to exit the cell. Here, we have used high-throughput quantitative proteomics combined with bioinformatics to characterize and quantify vesicle-mediated protein release from β-glucan-stimulated human macrophages differentiated in vitro from primary blood monocytes. We show that β-glucan stimulation induces vesicle-mediated protein secretion. Proteomic study identified 540 distinct proteins from the vesicles, and the identified proteins show a proteomic signature characteristic for their cellular origin. Importantly, we identified several receptors, including cation-dependent mannose6-phosphate receptor, macrophage scavenger receptor, and P2X7 receptor, that have not been identified from vesicles before. Proteomic data together with detailed pathway and network analysis showed that integrins and their cytoplasmic cargo proteins are highly abundant in extracellular vesicles released upon β-glucan stimulation. In conclusion, the present data provides a solid basis for further studies on the functional role of vesicular protein secretion upon fungal infection. KEYWORDS: macrophage, β-glucan, extracellular vesicle, quantitative proteomics, Dectin-1



α-mannans resulting in the activation of host defense against fungal infection.5 Likewise, the mannose receptor also recognizes mannans and directs fungal antigens into the endocytic pathway thereby promoting antigen presentation.6,7 During stimulatory conditions of infection and colonization, besides the intensive intracellular response, human cells actively communicate with each other by secreting various biomolecules including nucleic acids, lipids, and proteins.8,9 The secreted proteins can be released through various mechanisms including classical, signal sequence-dependent secretion and nonclassical, vesicle-mediated mechanisms. The vesicle-mediated protein secretion mechanisms include diverse types of membrane vesicles of endosomal and plasma membrane origin, which are collectively called “extracellular vesicles” (EVs).10 Exosomes are 40−100 nm vesicles that are thought to originate from multivesicular bodies (MVBs), which in turn arise in the endosomal pathway by inward budding of endosomal membrane.11 Exosomes are known to mediate secretion of many cellular proteins originating from various cell compartments. Their role has been primarily attributed to intercellular

INTRODUCTION Fungal infections (mycoses) are common diseases of varying severity that are caused by exposition to fungal yeast spores and hyphae, most often through inhalation or skin colonization. Fungal infections have attracted growing concern over recent years, as they are an imposing threat for patients undergoing prolonged hospitalizations and are often associated with immunosuppressive therapies, HIV infections, and medical intervention.1 Activation of the innate immune system depends on the recognition of pathogen-associated molecular patterns (PAMPs) by pattern recognition receptors (PRRs) of the innate immune cells, including dendritic cells and macrophages. The knowledge about innate immune reaction to fungal infections has increased rapidly after the discovery of specific fungi-recognizing receptors and their ligands. Fungal cell components, including β-glucans and α-mannas, are important PAMPs that are recognized by C-type lectin receptors (CLRs). Dectin-1, a family member of the CLRs, is the best characterized PRR for β-glucans (reviewed by Romani in 20112). In addition to Dectin-1, scavenger receptors CD36 and SCARF13 as well as complement receptor CR34 have been implicated as receptors for β-glucans. Also, Dectin-2 and Dectin-3 are important PRRs that are involved in recognition of © 2014 American Chemical Society

Received: December 17, 2013 Published: March 26, 2014 2468

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477

Journal of Proteome Research

Article

signaling, mainly in communication between immune cells.12 Moreover, exosomes have been shown to be taken up by macrophages,13 demonstrating their additional communication potential among cells. Quantitative proteomic analysis of EVs combined with bioinformatics is a powerful tool to gain in-depth knowledge of the protein composition of EVs and can shed new light on signaling roles that vesicles play in protein secretion and intercellular communication in innate immune system combating fungal infection. Here, we have used isobaric tag for relative and absolute quantitation labeling (iTRAQ) combined with high-throughput mass spectrometry14 and bioinformatics to characterize EVs released from human peripheral blood monocyte-derived macrophages stimulated with curdlan, a linear (1,3)-β-glucan that specifically activates PRR Dectin-1.



resuspended in 5 mL of fresh PBS. The tubes were centrifuged at 100 000g for additional 1 h. The pellets were resuspended in 100 μL of PBS and the total protein precipitation was performed directly using 2D Clean-Up Kit (GE Healthcare) according to the manufacturer’s instructions for mass spectrometry. For WB experiments, 10−15 mL of control and curdlan media was isolated and either treated as described above or centrifuged at 500g for 10 min and at 3000g for 30 min to remove cells and cell debris, directly transferred to 38.5 mL polyallomer tubes (Beckman), filled up with microvesicledepleted PBS, and centrifuged for 1 h at 100 000g using SW 28 Ti swinging bucket rotor (Beckman Coulter). The resulting pellet was washed with fresh PBS and centrifuged again at 100 000g. The supernatant was removed and the pellet was solubilized in 25 μL of Laemmli sample buffer. For electron microscopy (EM), washed and pelleted microvesicles were resuspended in 50 μL of PBS and fixed by 1:1 dilution with 4% paraformaldehyde of pH 7. The fixed vesicles were stored at 4 °C and loaded on EM grids within 1 week.

MATERIALS AND METHODS

Reagents

The antibodies against annexin 1 (sc-12740), alix (sc-53540), tsg101 (sc-7964), P2X7 receptor (sc-25698), Syk (sc-73087), cSrc (sc-130124), pan14-3-3 (sc-629) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The antibody against tubulin α/β (#2148) was purchased from Cell Signaling Technology (Beverly, MA). The antibody against CD11c (ITGAX, ab52632) was purchased from Abcam PLC (Cambridge, U. K.). Rabbit anti-IL-1β antibody has been previously described.15 Nonpyrogenic curdlan (Dako, Glostrup, Denmark) was suspended in sterile PBS at 1 mg/mL under mild heating and agitation, resulting in homogeneous colloidal suspension.

Electron Microscopy of Vesicles

The EM of vesicles was performed as described by Théry et al.16 with minor modifications. Briefly, 5 μL of fixed exosome suspension was transferred to Pioloform-carbon-coated golden grids and allowed to absorb for 20 min. The grids were subsequently washed and contrasted with uranyl acetate, embedded in the mixture of uranyl acetate and methyl cellulose on ice for 10 min, and air-dried. For immunogold labeling, fixed microvesicles were absorbed on the Pioloform-carbon-coated copper grids, and the grids were subsequently incubated with a 20 μg/mL solution of antibodies against annexin 1 or pan14-33. After washing in PBS, the grids were incubated in an appropriate secondary antibody conjugated to 10 nm gold particles and contrasted and embedded as mentioned above. Vesicles were observed at 80 kV with Jeol 1200 EX II transmission electron microscope.

Monocyte-Derived Macrophage Cultures and Stimulations

Peripheral blood mononuclear cells were isolated from leukocyte-rich buffy coat obtained from Finnish Red Cross Blood Transfusion Service and differentiated into macrophages by maintenance in Macrophage-SFM medium (GIBCO, Invitrogen, Carlsbad, CA) supplemented with 10 ng/mL GM-CSF (Biosource International) and antibiotics as described before.15 The cells were seeded at 1.4 million cells per well in 1 mL of medium. On the sixth day, the cells were washed two times with PBS, supplied with fresh RPMI 1640 medium (GIBCO) supplemented with L-glutamate and antibiotics, incubated at 37 °C with 5% CO2 for 1 h, and stimulated with 10 μg/mL curdlan or left untreated for 24 h. Where indicated, the cells were stimulated with 1 μg/mL LPS or 107/ mL heat-killed C. albicans (Invivogen, San Diego, CA).

Nanoparticle Tracking Analysis

After ultracentrifugation, EVs were resuspended in PBS and analyzed with nanoparticle tracking analysis (NTA) instrument LM14C with blue laser (405 nm, 60 mW, NanoSight Technology, London, U. K.) and CMOS camera (Hamamatsu Photonics K. K., Hamamatsu City, Japan) to determine the vesicle size distributions and concentrations. Samples were injected manually and data acquisition was done at ambient temperature. Samples were run in triplicates. Settings for data acquisition were as follows: basic, camera level 14, auto settings off, polydispersity and reproducibility high with particles per image 40−100 (acquisition time 90 s). Data was analyzed with NTA 2.3 software with settings expert, background extraction/ auto blur/autominimum track length on and minimum expected particle size 50 nm. Samples from two independent biological replicates were measured.

Isolation of Microvesicles

For iTRAQ experiments, 120 mL of conditioned culture media from curdlan-treated and control cells were collected and centrifuged at 500g for 10 min to remove suspended cells and further at 3000g for 30 min to remove residual cell debris. The supernatants were concentrated with Amicon 100 kDa MWCO centrifugal filter units (Millipore) and the concentrated medium was transferred to a fresh 5 mL ultracentrifugal polyallomer tube (Beckman). The filter membrane was washed three times with an additional 1 mL of sterile-filtered PBS previously deprived of contaminating microvesicles by 1 h ultracentrifugation at 100 000g and the liquid was added to the ultracentrifuge tube. The tube was filled to 5 mL with PBS and centrifuged for 1 h at 4 °C at 100 000g in SW 55 Ti swinging bucket rotor (Beckman Coulter). The supernatant was removed gently with a pipet and the microvesicle-pellets were

Proteomic and Bioinformatic Analysis

The protein pellets after the 2D Clean-Up Kit were dissolved in 20 μL of iTRAQ dissolution buffer, the proteins were reduced, alkylated, and digested using trypsin (Promega), and the resulting peptides labeled with 4plex iTRAQ Reagents Multiplex Kit (Applied Biosystems) according to the manufacturer’s instructions. After labeling, the samples were pooled and dried and the peptides were fractionated with 2469

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477

Journal of Proteome Research

Article

Figure 1. Characterization of extracellular vesicles released from human macrophages upon β-glucan stimulation. (A) Electron microscopy images of purified extracellular vesicles and immunogold labeling of exosomal marker proteins, pan14-3-3 and annexin A1. (B) Western blot analysis of exosomal marker proteins and silver stained SDS-PAGE from isolated vesicles. (C) Nanoparticle tracking analysis of EVs.

>1.3). Because the quantification was aiming to determine the total relative protein secretion in the vesicles, the automatic bias correction was turned off during search. All protein identification and quantification results were manually verified to ensure the following: (1) for each identified protein, at least two unique peptides with good quality MS/MS data were required, (2) all fragmentation spectra with both reporter ion peak height below 10 counts were removed from quantification results. Protein identification and relative quantitation data from the iTRAQ experiments are shown in Supporting Information Table 1. False discovery rates were determined using a concatenated regular and reversed sequence database as described by Elias et al.19 The FDR rates determined for three biological replicates were 0, 1.84, and 1.06%, respectively. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (http://proteomecentral. proteomexchange.org) via the PRIDE partner repository20 with the data set identifier PXD000642. All the identified proteins were manually compared with Exocarta21,22 and checked for the presence of signal peptide using SignalP 4.1 web-based server.23 Gene ontology (GO) analysis was done with Web-based Gene Set Analysis Toolkit (WebGestalt24,25) and the biological network, and pathway analysis was performed using Ingenuity Pathway Analysis software (IPA, Ingenuity Systems, Mountain View, CA, www. ingenuity.com).

strong cation exchange chromatography (SCX) using a PolySULPHOETHYL A 200 × 2.1 mm column (202SE0502; PolyLC INC.) connected to an Ä KTA HPLC system (Amersham Biosciences) as previously described.17 Each SCX fraction containing iTRAQ labeled peptides was analyzed twice using an Ultimate 3000 nanoliquid chromatography (LC) (Dionex) coupled to a QSTAR Elite hybrid quadrupole timeof-flight-mass spectrometer (AB Sciex) with nanoelectrospray ionization. The LC−MS/MS samples were first loaded on a ProteCol C18 trap column (SGE, 10 × 150 μm, 3 μm, 120 Å), followed by peptide separation on a PepMap100 C18 analytical column (LC Packings/Dionex, 15 × 75 μm, 5 μm, 100 Å) at 200 nL/min. The separation gradient consisted of 0−50% B in 120 min, 50% B for 3 min, 50−100% B in 2 min, and 100% B for 3 min (buffer A: 0.1% formic acid; buffer B: 0.08% formic acid in 80% acetonitrile). MS data were acquired automatically using Analyst QS 2.0 software. Information-dependent acquisition method consisted of a 0.5 s TOF-MS survey scan of m/z 400−1400. From every survey scan, the two most abundant ions with charge states +2 to +4 were selected for product ion scans, and each selected target ion was dynamically excluded for 60 s. Smart IDA was activated with automatic collision energy and automatic MS/MS accumulation.17,18 Protein identification and relative quantification was done with Protein Pilot 4.0 software (AB Sciex) using the Paragon search engine against the SwissProt database dated February 2012. The data from both technical replicates was processed together. The search criteria were cysteine alkylation with iodoacetamide, digestion enzyme: trypsin, biological modifications allowed, thorough search effort, detected protein threshold ensuring 95% confidence (unused protein score

SDS-PAGE and Western blotting

The samples were incubated at 95 °C for 5 min and an equal volume (5 to 7 μL) of each were separated on 12% SDS-PAGE, transferred to Immobilon-P Transfer Membranes (Millipore), 2470

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477

Journal of Proteome Research

Article

Figure 2. Quantitative proteomics of extracellular vesicles and classification of all the identified proteins. (A) Workflow for quantitative proteomics and summary of identification and quantification results. (B) Gene Ontology analysis based on cellular component and molecular function. (C) Venn diagram of the proteins identified having a predicted N-terminal signal peptide or being known exosomal proteins.

isolated from equal volumes of culture supernatant were compared, exosomal marker proteins were enriched in curdlanstimulated samples as compared to the controls; additionally, the total protein amount in the vesicle-fraction was higher in curdlan-stimulated samples, as visualized with silver staining (Figure 1B). Then we performed nanoparticle tracking analysis to determine the overall yield and particle size of the vesicles. This showed that curdlan-stimulated cells secrete over 3-fold more vesicles per milliliter compared with unstimulated cells, but the particle size remains similar in both samples (Figure 1C). Next, we performed quantitative proteomic analysis of EVs from β-glucan-stimulated human macrophages using iTRAQ labeling coupled with LC−MS/MS analysis. The proteomic analysis was done for three independent biological replicates. Altogether, we identified 540 distinct proteins, of which 145 proteins were reliably quantified (Figure 2A and Supporting Information Table 1). Also the iTRAQ quantification values unambiguously indicated robust increase in the secretion of vesicle-associated proteins (Supporting Information Table 1). All the identified proteins were classified based on their cellular component and molecular function (Figure 2B). This showed that membrane annotation was accurate for more than 60% of proteins (335) following annotations to macromolecular complex (217), cytosol (193), and nucleus (130). In agreement with earlier studies on microvesicles, cytoskeletal proteins (117) and endomembrane system proteins (94) also constitute an important fraction of identified proteins. The molecular function analysis revealed that vast majority (358) of the

blocked with 5% nonfat milk solution in TBS-Tween (TBS-T), and incubated overnight at 4 °C with primary antibodies for exosomal marker proteins and other described proteins of interest. The membranes were washed three times with TBS-T and incubated with the appropriate HRP-conjugated secondary antibody for 1 h at room temperature and, after that, washed with TBS-T and visualized with Western Lightning ECL (PerkinElmer, Waltham, MA).



RESULTS

Quantitative Proteome Analysis of Extracellular Vesicles Released from Human Macrophages upon β-Glucan Stimulation

In the present report, we have used high-throughput quantitative proteomics combined with bioinformatics to characterize and quantify extracellular vesicle-mediated protein release from β-glucan-stimulated human macrophages differentiated in vitro from primary monocytes. First, we examined the composition and homogeneity of the purified vesicles using EM. This shows that the preparation contained a mixture of vesicles whose diameter varied from 30 to 100 nm (Figure 1A), which is in agreement with previous reports on exosomes of mammalian origin. Occasionally also bigger vesicles were observed. The presence of exosomal markers 14-3-3β protein and annexin A1 was also visualized on the vesicles using immunogold labeling (Figure 1A, center and right panel). Next, the isolated vesicles were subjected to Western blotting to detect the well-known exosomal marker proteins annexin A1, alix, tsg101, and tubulin. When vesicles 2471

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477

Journal of Proteome Research

Article

Table 1. Macrophage signature membrane proteins are present on microvesicles isolated from human primary macrophages.a UniProt Accession # P11215 P05107 P21757 P08575 P15144 Q07954 P16671 P02786 O15031 P20702 P30273 P25705 Q13740 P08195 P05556 P05023 P54709 P08648 Q9NZM1 P05362 P53396 P78324

Macrophage signature membrane proteins integrin α M integrin β 2 macrophage mannose receptor 1 receptor-type tyrosine-protein phosphatase C aminopeptidase N prolow-density lipoprotein receptor-related protein 1 platelet glycoprotein 4 (CD36) transferrin receptor protein 1 plexin B2 integrin α X high-affinity immunoglobulin γ Fc receptor I ATP synthase subunit α, mitochondrial CD166 antigen (ALCAM) 4F2 cell-surface antigen heavy chain integrin β 1 sodium/potassium-transporting ATPase subunit α-1 sodium/potassium-transporting ATPase subunit β-3 integrin α-5 myoferlin intercellular adhesion molecule 1 ATP-citrate synthase tyrosine-protein phosphatase-n-receptor type substrate 1

ExoCarta + + + + + + +

+ + + + + + + + + + +

a

A total of 22 proteins with close human homologs reported as highly abundant proteins in the membrane proteome of differentiated mouse macrophages26 were identified with high confidence from microvesicles released from human primary macrophages.

proteins possesses protein binding activity. Other molecular functions assigned to the proteins suggest their catalytic activity (hydrolase activity (119)), transporter activity (61), molecular transducer activity (51), and structural function (67). The identified proteins were next analyzed using SignalP to determine the presence of the signal peptide sequence needed for conventional protein secretion and compared to ExoCarta, which is a database containing known exosomal proteins.21,22 From the proteins identified in our study, 109 had signal sequence and 431 have been reported in human ExoCarta (Figure 2C and Supporting Information Table 1). Among the proteins identified in this study that have not been reported in ExoCarta, we identified several immunity-related signaling proteins and transmembrane proteins. We also identified multiple leukocyte-associated proteins (HLA-DPA1, HLA-F, and other HLA histocompatibility complex members). Interestingly, EVs were enriched in immunity-related receptors, which have not previously been detected on exosomes. These include high-affinity IgE receptor γ chain (FCER1G), C5a anaphylatoxin chemotactic receptor (C5AR1), cation-dependent mannose-6-phosphate receptor (M6PR), macrophage scavenger receptor (MSR1), and P2X7 receptor (P2RX7). Vesicles originate from membrane structures, and most abundant proteins identified on them are membrane-associated. We compared all the proteins identified in our experiments to the recently published list of highly abundant mouse macrophage plasma membrane proteins.26 This shows that 17 most abundant membrane proteins identified by Becker et al. were

Figure 3. Canonical pathway and network analysis of EV proteome from human macrophages upon β-glucan stimulation. (A) Top canonical pathways of EV proteome based on all the identified proteins. (B) Top score networks based on the identified and quantified proteins. (C) Detailed view of network 1 demonstrating relations between numerous integrins, including ITGAM and ITGAX, which have not previously been identified on exosomes (bold and red). Insert: Western blot analysis of ITGAX present on EV fraction.

identified also in our proteomic analysis of macrophage-derived EVs. Additionally, five other highly abundant proteins were also identified (Table 1). Eight of these proteins are receptors that typically associate with the macrophage plasma membrane and participate in immune response, and four of them were not known to associate with human exosomes before. This data indicates that macrophage-derived EVs carry a proteomic signature characteristic for their cellular origin. Integrins and Their Cytoplasmic Cargo Proteins Are Highly Abundant in Extracellular Vesicles Released upon β-Glucan Stimulation

To get biological insight into the proteome composition of the vesicles, we performed pathway and network analysis with IPA 2472

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477

Journal of Proteome Research

Article

Figure 4. Integrins and their cytoplasmic cargo proteins are highly abundant in extracellular vesicles released upon β-glucan stimulation. (A) Identified integrins from EVs. (B) Classical integrin signaling pathways linking extracellular signal to cytoskeletal proteins. Proteins identified in EV proteome highlighted. (C) Western blot analysis of EV marker proteins and selected integrin-associated cytoskeleton-related proteins on isolated EVs after stimulation with β-glucan curdlan, heat-killed C. albicans (HKCA), or bacterial lipopolysaccharide (LPS).

identified altogether nine distinct integrin-subunits from EVs released upon β-glucan stimulation form macrophages, including three β-subunits and six α-subunits (Figure 4A). From these, two α-subunits, ITGAM and ITGAX, have not been previously identified from human EVs. The presence of ITGAX on microvesicles was also confirmed by Western blotting (Figure 3C, insert). Multiple integrin-associated cytoskeleton-related proteins, such as actin, zyxin, vinculin, and talin, as well as integrin-linked kinase ILK were also identified throughout proteomic experiments (Figure 4B and Supporting Information Table 1), further suggesting a functional role of integrin in EVs. To test whether the increase in vesicular release of these proteins is attributed specifically to stimulation of macrophages with PAMP of fungal origin, we next stimulated macrophages either with β-glucan curdlan, heat-killed C. albicans (HKCA) yeast, or bacterial lipopolysaccharide (LPS) and isolated the EVs. Western blot analyses on the EV proteins showed that the exosomal marker proteins TSG101 and Alix were present in all the EVs but clearly were more abundant after curdlan stimulation compared to HKCA and LPS stimulation (Figure 4C). Also, the secretion of ITGAX, α-actinin, talin, and ILK was more pronounced after curdlan stimulation compared to HKCA and LPS. In line with this, NTA analysis showed that curdlan-stimulated cells secrete slightly more vesicles compared to HKCA and clearly more compared to LPS-stimulated cells (data not shown).

software. Canonical pathway analysis of all identified proteins indicated a number of over-represented pathways (Figure 3A). Most of the top-ranked pathways are typical for immunityrelated cells. In particular, Fc-γ receptor-mediated phagocytosis and leukocyte extravasation signaling are important pathways of the antifungal innate immunity. Fc-γ receptor is involved in phagocytosis of fungal particles as well as in mediating Dectin2-induced signaling (reviewed by Romani in 20112). In addition, integrin signaling plays a vital role in inflammatory response, and integrin signaling was the most relevant signaling pathway represented in our proteomic data (Figure 3A). In order to analyze the biological impact of vesicular proteins whose release is regulated by Dectin-1 in more detail, we performed network analysis on 145 proteins with significant iTRAQ quantification values. This analysis identified six significant networks with 12−16 proteins in focus (Figure 3B), and network 1, “cellular movement, haematological system development and function, immune cell trafficking”, is also directly linked to integrin signaling (Figure 3C). Integrins are heterodimers of α- and β-subunits and combinations of different subunits give numerous different integrin receptors with specific binding properties. The integrins trigger a variety of intracellular signaling cascades, frequently linking the extracellular signal to cytoskeleton rearrangement. Upon extracellular stimulation, integrins undergo a conformational change (integrin activation), which allows the recruitment of several cytoplasmic proteins intracellular domain of integrins resulting in the formation of a large protein complex that interacts with the actin cytoskeleton.27 We 2473

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477

Journal of Proteome Research

Article

Figure 5. Extracellular vesicles released from human macrophages upon β-glucan stimulation contain P2X7 receptor and Syk and Src kinases as well as proIL-1β. (A) Western blot analysis of P2X7, Syk, Src, and IL-1β from the purified microvesicles as well as from the flow-through fraction containing proteins that are not secreted in vesicles. (B) Network analysis showing association between P2X7 receptor, NFκB, and CD cluster.

P2X7 Receptor and the Precursor Form of Interleukin-1β Are Secreted in Microvesicles

secreted in vesicles, we performed Western blot analysis with anti-IL-1β antibody that detects both the precursor and the mature form of the cytokine. Our results show that the precursor form of IL-1β is present in EVs isolated from curdlanstimulated macrophages. In contrast, the mature 17 kDa form of IL-1β was secreted independent of EVs (Figure 5A). Our data demonstrates that precursor and mature forms of IL-1β are released from human macrophages through different pathways.

Previous studies have demonstrated that Syk and Src tyrosine kinases are important components of Dectin-1 signaling.28 Our proteomic data indicated the presence of these tyrosine kinases in vesicles, and Western blot analysis confirmed this (Supporting Information Table 1 and Figure 5A), suggesting that macrophages exposed to fungal PAMPs can release signaling components to neighboring cells and, thereby, enhance their activation potential. Distant intercellular receptor transfer by exosomes has been demonstrated to play a role in exosome signaling.29 Our proteomic data revealed that purinergic receptor P2RX7 is secreted in EVs from β-glucan stimulated macrophages, and Western blot data confirms this (Supporting Information Table 1 and Figure 5A). P2RX7 is a ligand-gated ion channel belonging to the P2X family of purinoceptors, which, once activated by ATP, induce large pore formation, increased flux of molecules through the membrane and ultimately, during prolonged activation, cell death. This receptor has also been shown to activate IL-1β maturation and secretion in both conventional and vesicle-mediated way.30 Network analysis included P2RX7 in one of the relationship networks (Figure 5B) with “cellular function and maintenance, molecular transport, and cell-to-cell signaling and interaction” as functions assigned by IPA. The protein is placed in the immediate neighborhood of dynamin, caveolin as well as VAMP3 and Rab5, assembling into cluster of proteins involved in caveolarmediated endocytosis. It has been previously shown that the P2X7 receptor can be internalized upon ATP stimulation.31 Exosomes are born in endosomes and our study demonstrates that the internalized P2X7 receptor can be nonclassically secreted in EVs from human macrophages. Ligation of P2X7 receptor is known to activate secretion of interleukin-1β (IL-1β), which is the most important proinflammatory cytokine produced by human macrophages. We have previously shown that Dectin-1 pathway activates caspase1-dependent proteolytic processing of precursor form of IL-1β to its biologically active form.32 To study whether IL-1β is



DISCUSSION Fungal infections occurrence has been causing increasing problems during prolonged or palliative hospitalization as well as in immunocompromised patients. Understanding the intracellular signaling cascades and intercellular signaling that mediates the immune reaction against fungal infection is critical to coming up with successful and reliable treatments. The innate immune system is the first line of defense against microbial infection and tissue damage. Macrophages are a central cell type of the innate immune system. They respond to activating stimuli by producing inflammatory mediators that are delivered to neighboring cells through multiple protein secretion pathways, including vesicle-mediated secretion. Here, we report a quantitative proteomic study combined with bioinformatic analysis of extracellular vesicles released from human macrophages upon stimulation with linear (1,3)-βglucan, curdlan, which activates the innate immune system through the Dectin-1 receptor. We analyzed three biological replicates, which resulted in the identification of 540 distinct proteins and reliable quantification of 145 proteins. We used macrophages differentiated from blood-derived human monocytes in our experiments, where the starting material is a limiting factor that is reflected by the quality of quantification data obtained. It can also be seen from the identification results where the numbers of identifications vary between the biological replicates. It must be noted, however, that the data obtained from primary cells is biologically much more relevant than what can be obtained using cell lines. Comparison of all the identified proteins to ExoCarta showed that approximately 80% of the identified proteins in 2474

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477

Journal of Proteome Research

Article

or ITGAX, which leads to formation of macrophage receptor 1 or iC3B receptor, respectively. These receptors are members of the complement system, whose role in β-glucan recognition was shown by Ross et al.,42 before discovery of Dectin-1. The former has been associated with phagocytosis of M. tuberculosis43 and the latter with recognition of C. albicans.40,44,45 The proteins identified in our study include also numerous receptors expressed on macrophages. Exosomal transfer of receptors has been recognized as important for the organism to combat the disseminating disease, but it has also been linked to the disease spreading itself.29 Interestingly, P2RX7 has been found to mediate MHC class II and IL-1β exosomal secretion in ATP-stimulated macrophages and dendritic cells in NLRP3/ ASC-dependent manner.46,47 To strengthen these links, previous studies in our group demonstrated that (1,3)-βglucans activate NLRP3 inflammasome in human primary macrophages, which is the central molecular complex involved in caspase-1 activation and subsequent processing of precursor form of IL-1β.32 The detection of precursor form of IL-1β, but not the mature form of the cytokine, in microvesicles is an intriguing finding. These results suggest that EVs play a role in intercellular proinflammatory signal spreading by mediating precursor form of IL-1β as well as P2X7 receptor to adjacent macrophages. If these macrophages encounter ATP, which is often released during tissue damage in microbial infections, caspase-1 is activated and the mature form of IL-1β is secreted. Taken together, our data show that β-glucan stimulation induces extracellular vesicle-mediated protein secretion from human macrophages, and these vesicles carry a set of proteins typical for their origin that might have a significant role in molecular transfer of biological information. It is tempting to speculate that in our model of fungal infection EVs play a specific role as “danger signals” delivered directly to target cells. There is a possibility that the high number of immunityassociated proteins in this way can be delivered to cells and by means of receptor or proinflammatory precursors transfer provide them with higher sensitivity to potential danger associated with spreading infection. A similar role for EVs in immune response and inflammatory diseases have also been suggested by two recent reviews.9,48 In conclusion, our current data provides an important background for further functional studies on extracellular vesicles released from human macrophages upon fungal infection.

our study have previously been found in exosomes. Interestingly, we identified 109 proteins that are predicted to be classically secreted (Supporting Information Table 1) However, most (82) of these are also found in ExoCarta. The presence of a signal peptide does not imply that the protein is secreted in the soluble form. Many, in fact, remain attached to the membranes either by spanning it (transmembrane proteins) or by lipid anchors. In our data, only 23 of signal-peptide-containing proteins are reported to be secreted to the extracellular space, whereas the others are membranespanning, attached via lipid anchor, or unknown (data not shown). Altogether, this data strengthens the contribution to identification of novel extracellular vesicle-associated proteins, which likely results from membrane-reassembly processes. Canonical pathway analysis of the identified proteins indicated the importance of different endo- and phagocytosis mechanisms as probable mechanisms sorting the proteins to secreted vesicles. Exosomes are known to originate from MVBs, which in turn gather endocytosed material; therefore, it is not surprising to see strong relation of identified proteins to endocytosis-related pathways. It is also likely that similar molecular machinery can be involved in both endo- and exocytosis, providing a link between the two processes. Given the high abundance of membrane components in the analyzed proteome, it is likely that membrane-shed microvesicles also significantly contribute to the overall unconventional protein secretion during the β-glucan challenge. Our quantitative proteome data together with pathway and network analysis clearly showed that integrins and intracellular proteins involved in integrin receptor aggregation and signaling upon extracellular matrix binding are highly abundant in EVs after β-glucan stimulation. Integrins are cell-surface receptors that mediate adhesion to cells and to the extracellular matrix. A common feature of many types of exosomes is their expression of adhesion molecules, like integrins, on their surface.33 The binding of secreted vesicles to the recipient cell is speculated to be mediated by interaction between exosomal integrins and ICAM-1 on recipient cell surface.34 Also integrin-mediated exosome binding to extracellular proteins such as fibronectin has been reported.35 After interaction, exosomes might fuse with recipient cell membrane and transfer receptors to cells.36−38 Alternatively, vesicles can be internalized by endocytosis. Integrins do not themselves possess a kinase domain or enzymatic activity, but the interaction with cytoplasmic proteins enables the unique ability to signal bidirectionally.39 In addition, the affinity of integrin to its ligands is strictly regulated by interaction with the actin cytoskeleton. It is a complex process involving multiple proteins and signaling pathways; however, the critical point is talin binding to the cytoplasmic tails of β-integrin and the activation of Rap1.27 Altogether, this analysis suggests that EVs from macrophages activated by β-glucans are equipped with molecular players that may facilitate their delivery and uptake by target cells; however, the functional roles remain to be elucidated. Integrin αX-subunit has recently been identified as a receptor participating in innate immune response of leukocytes against C. albicans.40 Identification of integrin αX on EVs released from β-glucan-activated macrophages could be characteristic for the fungal infection. Integrin αM likewise has been linked to the immune reaction and claimed a role in phagocytosis and cellular activation, as well as in complement system.41 Network analysis showed the association of ITGB2 with either ITGAM



ASSOCIATED CONTENT

S Supporting Information *

Protein identification and quantification results. All proteins identified and quantified in three biological replicates iTRAQ experiments are shown together with their relative quantification (curdlan:control) ratios. The statistically significant (p < 0.05) quantifications are indicated in bold. ExoCarta status of all identified proteins (Sep 2012) is indicated in column J. Predicted signal peptide presence is indicated in column K. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*T. A. Nyman. Phone: +358 9 191 59411. Fax +358 9 191 59930. E-mail: tuula.nyman@helsinki.fi. Address: Institute of Biotechnology, P.O. Box 65 (Viikinkaari 1), FI-00014 University of Helsinki, Finland. 2475

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477

Journal of Proteome Research

Article

Notes

Saccharomyces cerevisiae using amine-reactive isobaric tagging reagents. Mol. Cell. Proteomics 2004, 3, 1154−69. (15) Pirhonen, J.; Sareneva, T.; Kurimoto, M.; Julkunen, I.; Matikainen, S. Virus infection activates IL-1 β and IL-18 production in human macrophages by a caspase-1-dependent pathway. J. Immunol. 1999, 162, 7322−9. (16) Théry, C.; Clayton, A.; Amigorena, S.; Raposo, G.Isolation and characterization of exosomes from cell culture supernatants and biological fluids. In Current Protocols in Cell Biology; Bonifacino, J. S., Dasson, M., Harford, J. B., Lippincott-Schwartz, J., Yamada, K. M., Eds.; Wiley: Hoboken, NJ, 2006; Unit 3.22, pp 1−29 (17) Lietzén, N.; Ö hman, T.; Rintahaka, J.; Julkunen, I.; Aittokallio, T.; Matikainen, S.; Nyman, T. A. Quantitative subcellular proteome and secretome profiling of influenza A virus-infected human primary macrophages. PLoS Pathog. 2011, 7, e1001340 DOI: 10.1371/ journal.ppat.1001340. (18) Ö hman, T.; Lietzén, N.; Välimäki, E.; Melchjorsen, J.; Matikainen, S.; Nyman, T. A. Cytosolic RNA recognition pathway activates 14−3-3 protein mediated signaling and caspase-dependent disruption of cytokeratin network in human keratinocytes. J. Proteome Res. 2010, 9, 1549−64. (19) Elias, J. E.; Gygi, S. P. Target-decoy search strategy for increased confidence in large-scale protein identifications by mass spectrometry. Nat. Methods 2007, 4, 207−14. (20) Vizcaíno, J. A.; Côté, R. G.; Csordas, A.; Dianes, J. A.; Fabregat, A.; Foster, J. M.; Griss, J.; Alpi, E.; Birim, M.; Contell, J.; O’Kelly, J.; Schoenegger, A.; Ovelleiro, D.; Pérez-Riverol, Y.; Reisinger, F.; Ríos, D.; Wang, R.; Hermjakob, H. The Proteomics Identifications (PRIDE) database and associated tools: status in 2013. Nucleic Acids Res. 2013, 41 (D1), D1063−9. (21) Mathivanan, S.; Simpson, R. J. ExoCarta: a compendium of exosomal proteins and RNA. Proteomics 2009, 9, 4997−5000. (22) Mathivanan, S.; Fahner, C. J.; Reis, G. E.; Simpson, R. J. ExoCarta 2012: database of exosomal proteins, RNA and lipids. Nucleic Acids Res. 2012, 40, D1241−4. (23) Nordahl, T.; Brunak, S.; von Heijne, G.; Nielsen, H. SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat. Methods 2011, 8, 785−786. (24) Zhang, B.; Kirov, S. A.; Snoddy, J. R. WebGestalt: an integrated system for exploring gene sets in various biological contexts. Nucleic Acids Res. 2005, 33, W741−8. (25) Wang, J.; Duncan, D.; Shi, Z.; Zhang, B. WEB-based GEne SeT AnaLysis Toolkit (WebGestalt): update 2013. Nucleic Acids Res. 2013, 41, W77−83. (26) Becker, L.; Liu, N.-C.; Averill, M. M.; Yuan, W.; Pamir, M.; Peng, Y.; Irwin, A. D.; Fu, X.; Bornfeldt, K. E.; Heinecke, J. W. Unique proteomic signature distinguish macrophages and dendritic cells. PLoS One 2012, 7, e33297 DOI: 10.1371/journal.pone.0033297. (27) Kim, C.; Ye, F.; Ginsberg, M. H. Regulation of intregrin activation. Annu. Rev. Cell Dev. Biol. 2011, 27, 321−45. (28) Underhill, D. M.; Rossnagle, E.; Lowell, C. A.; Simmons, R. M. Dectin-1 activates Syk tyrosine kinase in a dynamic subset of macrophages for reactive oxygen production. Blood 2005, 106, 2543−2550. (29) Lee, Y.; Andaloussi, S. E. L.; Wood, M. J. A. Exosomes and microvesicles: extracellular vesicles for genetic information transfer and gene therapy. Hum. Mol. Genet. 2012, 21, R125−34, DOI: 10.1093/ hmg/dds317. (30) MacKenzie, A.; Wilson, A. L.; Kiss-Toth, E.; Dower, S. K.; North, R. A.; Suprenant, A. Rapid secretion of interleukin-1β by microvesicle shedding. Immunity 2001, 15, 825−35. (31) Feng, Y. H.; Wang, L.; Wang, Q.; Li, X.; Zeng, R.; Gorodeski, G. I. ATP stimulates GRK-3 phosphorylation and β-arrestin-2-dependent internalization of P2X7 receptor. Am. J. Physiol.: Cell Physiol. 2005, 288, C1342−56, DOI: 10.1152/ajpcell.00315.2004. (32) Kankkunen, P.; Teirilä, L.; Rintahaka, J.; Alenius, H.; Wolff, H.; Matikainen, S. 1,3)-β-glucans activate both Dectin-1 and NLRP3 inflammasome in human macrophages. J. Immunol. 2010, 184, 6335− 42.

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the Academy of Finland, grants 135628, 140950, and 272931; Sigrid Jusélius Foundation and Viikki Doctoral Program in Molecular Biosciences. The authors would like to thank Dr Pia Siljander and Maria Aatonen from the Department of Biochemistry, University of Helsinki, for the help with NTA analysis.



REFERENCES

(1) Auberger, J.; Lass-Flörl, C.; Aigner, M.; Clausen, J.; Gastl, G.; Nachbaur, D. Invasive fungal breakthrough infections, fungal colonization and emergence of resistant strains in high-risk patients receiving antifungal prophylaxis with posaconazole: real-life data from a single-centre institutional retrospective observational study. J. Antimicrob. Chemother. 2012, 67 (9), 2268−73. (2) Romani, L. Immunity to fungal infections. Nat. Rev. Immunol. 2011, 11, 275−88. (3) Means, T. K.; Mylonakis, E.; Tampakakis, E.; Colvin, R. A.; Seung, E.; Puckett, L.; Tai, M. F.; Stewart, C. R.; Pukkila-Worley, R.; Hickman, S. E.; Moore, K. J.; Calderwood, S. B.; Hacohen, N.; Luster, A. D.; Khoury, J. E. Evolutionarily conserved recognition and innate immunity to fungal pathogens by the scavenger receptors SCARF1 and CD36. J. Exp Med. 2009, 206 (3), 637−53. (4) Thornton, B. P.; Vetvicka, V.; Pitman, M.; Goldman, R. C.; Ross, G. D. Analysis of the sugar specificity and molecular location of βglucan-binding lectin site of complement receptor type 3 (CD11D/ CD18). J. Immunol. 1996, 156, 1235−46. (5) Zhu, L.-L.; Zhao, X.-Q.; Jiang, C.; You, Y.; Chen, X.-P.; Jiang, Y.Y.; Jia, X.-M.; Lin, X. C-type lectin receptors dectin-3 and dectin-2 form a heterodimeric pattern-recognition receptor for defense against fungal infection. Immunity 2013, 39, 324−34. (6) Lam, J. S.; Huang, H.; Levitz, S. M. Effect of differential N-linked and O-linked mannosylation on recognition of fungal antigens by dendritic cells. PLoS One 2007, 2, e33297 DOI: 10.1371/journal.pone.0001009. (7) Cambi, A.; Netea, M. G.; Mora-Montes, H. M.; Gow, N. A. R.; Hato, S. V.; Lowman, D. W.; Kullberg, B.-J.; Torensma, R.; Williams, D. L.; Figdor, C. G. Dendritic cell interaction with Candida albicans critically depends on N-linked mannan. J. Biol. Chem. 2008, 283, 20590−9. (8) Record, M.; Carayon, K.; Poirot, M.; Silvente-Poirot, S. Exosomes as new vesicular lipid transporters involved in cell-cell communication and various pathophysiologies. Biochim. Biophys. Acta, Mol. Cell Biol. Lipids 2014, 1841, 108−20. (9) Robbins, P. D.; Morelli, A. E. Regulation of immune responses by extracellular vesicles. Nat. Rev. Immunol. 2014, 14, 195−208. (10) Raposo, G.; Stoorvogel, W. Extracellular vesicles: exosomes, microvesicles and friends. J. Cell. Biol. 2013, 200, 373−83. (11) Pan, B. T.; Teng, K.; Wu, C.; Adam, M.; Johnstone, R. M. Electron microscopic evidence for externalization of the transferring receptor in vesicular form in sheep reticulocytes. J. Cell. Biol. 1985, 101, 942−8. (12) Chaput, N.; Théry, C. Exosomes: immune properties and potential clinical implementations. Semin. Immunopathol. 2011, 33, 419−40. (13) Lässer, C.; Aikhani, V. S.; Ekström, K.; Eldh, M.; Parredes, P. T.; Bossios, A.; Sjörstrand, M.; Gabrielsson, S.; Lötvall, J.; Valadi, H. Human saliva, plasma and breast milk exosomes contain RNA: uptake by macrophages. J. Transl. Med. 2011, 9, 1−8. (14) Ross, P. L.; Huang, Y. N.; Marchese, J. N.; Williamson, B.; Parker, K.; Hattan, S.; Khainovski, N.; Pillai, S.; Dey, S.; Daniels, S.; Purkayastha, S.; Juhasz, P.; Martin, S.; Bartlet-Jones, M.; He, F.; Jacobson, A.; Pappin, D. J. Multiplexed protein quantitation in 2476

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477

Journal of Proteome Research

Article

(33) Théry, C.; Ostrowski, M.; Segura, E. Membrane vesicles as conveyors of immune responses. Nat. Rev. Immunol. 2009, 9, 581−93. (34) Clayton, A.; Turkes, A.; Dewitt, S.; Steadman, R.; Mason, M. D.; Hallett, M. B. Adhesion and signaling by B cell-derived exosomes: the role of integrins. FASEB J. 2004, 18, 977−9. (35) Rieu, S.; Géminard, C.; Rabesandratana, H.; Sainte-Marie, J.; Vidal, M. Exosomes released during reticulocyte maturation bind to fibronectin via integrin α4β1. Eur. J. Biochem. 2000, 267, 583−90. (36) Skog, J.; Würdinger, T.; van Rijn, S.; Meijer, D. H.; Gainche, L.; Curry, W. T.; Carter, B. S.; Krichevsky, A. M.; Breakefield, X. O. Glioblastoma microvesicles transport RNA and proteins that promote tumour growth and provide diagnostic biomarkers. Nat. Cell Biol. 2008, 10, 1470−6. (37) Valadi, H.; Ekström, K.; Bossios, A.; Sjöstrand, M.; Lee, J. J.; Lötvall, J. O. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat. Cell Biol. 2007, 9, 654−9. (38) Al-Nedawi, K.; Meehan, B.; Micallef, J.; Lhotak, V.; May, L.; Guha, A.; Rak, J. Intercellular transfer of the oncogenic receptor EGFRvIII by microvesicles derived from tumour cells. Nat. Cell Biol. 2008, 10, 619−24. (39) Bouvard, D.; Pouwels, J.; De Franceschi, N.; Ivaska, J. Integrin inactivators: balancing cellular functions in vitro and in vivo. Nat. Rev. Mol. Cell Biol. 2013, 14, 430−42. (40) Jawhara, S.; Pluskota, E.; Verbovetskiy, D.; SkomorovskaProkvolit, O.; Plow, E. F.; Soloview, D. A. Integrin αXβ2 is a leukocyte receptor for Candida albicans and is essential for protection against fungal infections. J. Immunol. 2012, 189, 2468−77. (41) Solovjov, D.; Pluskota, E.; Plow, E. Distinct roles for the α and β subunits in the functions of integrin αMβ2. J. Biol. Chem. 2005, 280, 1336−45. (42) Ross, G. D.; Cain, J. A.; Myones, B. L.; Newman, S. L.; Lachmann, P. J. Specificity of membrane complement receptor type three (CR3) for β-glucans. Complement 1987, 4, 61−74. (43) Schlesinger, L. S.; Bellinger-Kawahara, C. G.; Payne, N. R.; Horwitz, M. A. Phagocytosis of Mycobacterium tuberculosis is mediated by human monocyte complement receptors and complement component C3. J. Immunol. 1990, 144, 2771−80. (44) Forsyth, C. B.; Mathews, H. L. Lymphocytes utilize CD11b/ CD18 for adhesion to Candida albicans. Cell. Immunol. 1996, 170, 91− 100. (45) Forsyth, C. B.; Plow, E. F.; Zhang, L. Interaction of the fungal pathogen Candida albicans with integrin CD11b/CD18: recognition by the I domain is modulated by the lectin-like domain and the CD18 subunit. J. Immunol. 1998, 161, 6198−205. (46) Qu, Y.; Franchi, L.; Nunez, G.; Dubyak, G. R. Nonclassical IL-1 β secretion stimulated by P2X7 receptors is dependent on inflammasome activation and correlated. J. Immunol. 2007, 179, 1913−25. (47) Qu, Y.; Ramachandra, L.; Mohr, S.; Franchi, L.; Harding, C. V.; Nunez, G.; Dubyak, G. R. P2X7 receptor-stimulated secretion of MHC class II-containing exosomes requires the ASC/NLRP3 inflammasome but is independent of caspase-1. J. Immunol. 2009, 182, 5052−62. (48) Buzas, E. I.; György, B.; Nagy, G.; Falus, A.; Gay, S. Emerging role of extracellular vesicles in inflammatory diseases. Nat. Rev. Rheumatol. 2014, No. 10.1038/nrrheum.2014.19.

2477

dx.doi.org/10.1021/pr4012552 | J. Proteome Res. 2014, 13, 2468−2477