Calpain Activity Is Essential for ATP-Driven Unconventional Vesicle

Home , Calcium signaling, Calpain

Calpain Activity Is Essential for ATP-Driven Unconventional Vesicle-Mediated Protein Secretion and Inflammasome Activation in Human Macrophages This information is current as of October 1, 2021. Elina Välimäki, Wojciech Cypryk, Juhani Virkanen, Katariina Nurmi, Pauli M. Turunen, Kari K. Eklund, Karl E. Åkerman, Tuula A. Nyman and Sampsa Matikainen J Immunol published online 16 September 2016

http://www.jimmunol.org/content/early/2016/09/15/jimmun Downloaded from ol.1501840

Supplementary http://www.jimmunol.org/content/suppl/2016/09/15/jimmunol.150184 Material 0.DCSupplemental http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

by guest on October 1, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published September 16, 2016, doi:10.4049/jimmunol.1501840 The Journal of Immunology

Calpain Activity Is Essential for ATP-Driven Unconventional Vesicle-Mediated Protein Secretion and Inﬂammasome Activation in Human Macrophages

Elina Va¨lima¨ki,*,† Wojciech Cypryk,* Juhani Virkanen,‡ Katariina Nurmi,x Pauli M. Turunen,{ Kari K. Eklund,x Karl E. A˚ kerman,{ Tuula A. Nyman,* and Sampsa Matikainen†,x

Extracellular ATP is an endogenous danger signal that is known to activate inflammatory responses in innate immune cells, including macrophages. Activated macrophages start to secrete proteins to induce an immune response, as well as to recruit other immune cells to the site of infection and tissue damage. In this study, we characterized the secretome (i.e., the global pattern of secreted proteins) of ATP-stimulated human macrophages. We show that ATP stimulation activates robust vesicle-mediated un- Downloaded from conventional protein secretion, including exosome release and membrane shedding, from human macrophages. Pathway analysis of the identified secreted proteins showed that calpain-related pathways were overrepresented in the secretome of ATP-stimulated cells. In accordance with this, calpains, which are calcium-dependent nonlysosomal cysteine proteases, were activated upon ATP stimulation through a P2X purinoceptor 7 receptor–dependent pathway. Functional studies demonstrated that calpain activity is essential for the P2X purinoceptor 7 receptor–mediated activation of unconventional protein secretion. Unconventional protein secretion was followed by cell necrosis and NLRP3 inflammasome–mediated secretion of the mature form of the proinflammatory http://www.jimmunol.org/ cytokine IL-1b. Furthermore, ATP-driven NLRP3 inflammasome activation was also dependent on calpain activity. Interestingly, pro–IL-1b and inflammasome components ASC and caspase-1 were released by ATP-activated macrophages through a vesicle- mediated secretion pathway. In conclusion, to our knowledge, we provide the first global characterization of proteins secreted by ATP-activated human macrophages and show a pivotal role for calpains in the activation of the inflammatory response during ATP exposure. The Journal of Immunology, 2016, 197: 000–000.

he innate immune system is activated during the early Extracellular ATP is a well-recognized damage-associated molec- response to microbial infection and tissue damage. Mac- ular pattern by the innate immune system. During normal homeostasis, T rophages are the central effector cells of innate immunity, extracellular ATP is rapidly hydrolyzed by ectonucleotidases; however, by guest on October 1, 2021 and their pattern recognition receptors detect the presence of pathogen- ATP released upon tissue damage exceeds the amount readily hy- associated molecular patterns and damage-associated molecular pat- drolyzed, resulting in the detection of ATP by tissue-resident mac- terns (1, 2). This recognition results in the activation of macrophages. rophages. The receptor for extracellular ATP is P2X purinoceptor They start to secrete cytokines to mount an inflammatory response, 7 (P2X7), which is an ATP-gated ion channel located on the plasma chemokines to recruit other immune cells to the site of infection or membrane of many cell types (4). Binding of ATP causes the channel inflammation, and other proteins that induce antimicrobial defense to dilate, allowing small cations to move across the plasma membrane and tissue regeneration (3). and resulting in changes in the ion balance of the cells. Activation of P2X7 initiates multiple, probably cell-type–specific downstream sig- *Institute of Biotechnology, University of Helsinki, 00014 Helsinki, Finland; naling events, such as reactive oxygen species formation, cell pro- †Finnish Institute of Occupational Health, 00250 Helsinki, Finland; ‡Department of x liferation, activation of transcription, and induction of apoptosis (5). Geosciences and Geography, University of Helsinki, 00014 Helsinki, Finland; Di- vision of Rheumatology, Helsinki University Hospital, Helsinki University, 00015 In macrophages, an important P2X7 downstream signaling event is Helsinki, Finland; and {Department of Physiology, Institute of Biomedicine, Biomedicum activation of the NLRP3 inflammasome, which is a protein complex Helsinki, University of Helsinki, 00014 Helsinki, Finland consisting of NLRP3, adaptor protein ASC, and cysteine protease ˚ ORCIDs: 0000-0002-3129-1678 (P.M.T.); 0000-0003-4625-3081 (K.E.A.). caspase-1 (6). Assembly of the NLRP3 inflammasome results in Received for publication August 20, 2015. Accepted for publication August 21, 2016. activation of caspase-1. Activated caspase-1 cleaves pro–IL-1b and This work was supported by Academy of Finland Grants 135628, 140950, 272931, pro–IL-18 to biologically active cytokines IL-1b and IL-18, which and 255826, the Sigrid Juse´lius Foundation, the National Doctoral Programme in Informational and Structural Biology (to E.V.), and the Integrative Life Science are secreted and initiate a strong inflammatory response (7, 8). The Doctoral Program (to W.C.). P2RX7 gene is highly polymorphic, and the activation of P2X7 is Address correspondence and reprint requests to Dr. Sampsa Matikainen, Division of associated with many inflammatory, immune, neurologic, and mus- Rheumatology, Helsinki University Hospital, Helsinki University, P.O. Box 63, culoskeletal diseases. 00014 Helsinki, Finland. E-mail address: sampsa.matikainen@helsinki.fi Calpains are calcium-dependent nonlysosomal cysteine prote- The online version of this article contains supplemental material. ases that are expressed ubiquitously in mammals and many other Abbreviations used in this article: BMDM, bone marrow–derived macrophage; DAVID, Database for Annotation, Visualization and Integrated Discovery; EV, extracellular ves- organisms. The calpain superfamily in humans consists of 18 genes icle; GO, Gene Ontology; ITGAX, integrin a-X; LC-MS/MS, liquid chromatography– coding for ubiquitously expressed and tissue-specific proteases (9). tandem mass spectrometry; LDH, lactate dehydrogenase; MSU, monosodium urate; Two ubiquitously expressed major species of calpains, calpain-1 NTA, nanoparticle tracking analysis; P2X7, P2X purinoceptor 7. (m-calpain) and calpain-2 (m-calpain), are the well-studied pro- Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 teases among the calpain family. They form heterodimers with

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1501840 2 CALPAINS AND ATP-DRIVEN PROTEIN SECRETION calpain small subunit-1 (also calpain-4), which is required for urate (MSU) was purchased from Enzo Life Sciences (Farmingdale, NY). catalytic activity of the enzyme. The activity of calpains is regulated The pharmacological inhibitors CA-074 Me (25 mM), MDL 28170 (Calpain by binding of Ca2+ and phospholipids and by phosphorylation. In Inhibitor III; 10–100 mM), PD 150606 (100 mM), calpeptin (100 mM), and Z-YVAD-FMK (Caspase-1 Inhibitor VI; 25–50 mM) (all from Calbiochem), addition, calpastatin is an endogenous calpain-specific inhibitor (10). as well as BAPTA-AM (10 mM; Enzo Life Sciences) and AZ 10606120 2+ Binding of Ca to calpain promotes translocation of the calpain (10 mM; Tocris Bioscience) were added to the cells 1 h prior to stimulation. heterodimer to cell membranes, after which both subunits are au- Secretome characterization tolyzed, and the enzyme becomes fully active and is released into cytosol (11). A well-known substrate for calpains is proinflammatory For secretome analysis, macrophages were stimulated in serum-free RPMI cytokine IL-1a, and the cleavage and secretion of IL-1a are de- 1640 (supplemented with L-glutamine and antibiotics; all from Life Technologies) with 3 mM ATP for 15 min, after which cell culture su- pendent on calpain activity (10, 12). However, there is little infor- pernatants were collected. Supernatants from cells of three donors were mation whether calpains regulate protein secretion in general. always pooled together to reduce individual variation among blood donors. Protein secretion allows communication between distant cells. Supernatants were concentrated with Amicon Ultra-15 centrifugal filter Secreted proteins control and coordinate many biological activities in devices with a cut-off of 10,000 nominal molecular weight limit. The multicellular organisms, such as growth, differentiation, apoptosis, concentrated supernatants were loaded onto 12% TGX mini gel (Bio-Rad), and proteins were separated at 200 V for 30 min. Subsequently, the gel was and immune responses (13). We (14–17) recently published several silver-stained in a mass spectrometry–compatible fashion (25). For iden- system-level characterizations of protein secretion by immune cells tification, the whole sample lanes were cut to 12 pieces of equal size, that highlight the importance of secretome analysis in the global followed by in-gel digestion with trypsin. The resulting peptides were characterization of the innate immune response. Protein secretion analyzed with an Ultimate 3000 nano-LC (Dionex) and a QSTAR Elite from cells is mediated through conventional and unconventional hybrid quadrupole/time-of-flight mass spectrometer (AB Sciex) with nano- electrospray ionization, as described (14, 26). Downloaded from pathways. Conventionally secreted proteins have a signal peptide on The obtained liquid chromatography–tandem mass spectrometry (LC-MS/ their N-terminal end that directs them through endoplasmic reticu- MS) data were searched with the in-house Mascot search engine version lum and Golgi apparatus to vesicles, which fuse with plasma mem- 2.4.0 through the ProteinPilot 4.5 interface (AB Sciex) against the database brane and release their cargo. Unconventionally secreted proteins Swiss-Prot (20200 Homo sapiens sequences). The following criteria for lack the signal peptide and are secreted directly through plasma Mascot searches were used: Homo sapiens as taxonomy; trypsin digestion with one missed cleavage allowed; carbamidomethyl modification of cys- membrane, via dedicated transporter channels, or via specific vesi- teine as a fixed modification; and oxidation of methionine and phosphory- cles, such as exosomes derived from multivesicular bodies (18, 19). lation of serine, threonine, and tyrosine as a variable modification. Peptide http://www.jimmunol.org/ Activation of P2X7 rapidly changes cell morphology and induces mass tolerance was 50 ppm, and fragment mass tolerance was 0.2 Da. trafficking and secretion of membrane-derived vesicles. Membrane Protein identifications that had probability-based Mascot MOWSE scores , blebbing (20), shedding of microvesicles (21), and release of $ 45 and p 0.05 were considered reliable high-confidence identifications. False discovery rates were calculated using a concatenated normal and re- phagolysosomes and exosomes from the cells were detected upon versed sequence database, a previously reported method (27), and were P2X7 activation (22). However, the molecular mechanisms that 2.1–3.2% in datasets. Protein identifications were done from three bio- regulate this membrane traffic have not been fully characterized. logical replicate samples for which a pool of cell culture supernatants from To our knowledge, this study provides the first in-depth secretome three donors was used as one sample. Proteins identified in at least two biological replicates were subjected to further bioinformatics analysis. characterization of human primary macrophages upon ATP stimu- The identified proteins were assigned to Gene Ontology (GO) (28) class by guest on October 1, 2021 lation. We show that ATP activates vesicle-mediated unconven- cellular component using GoMiner (29). Query parameters were UniProt, tional protein secretion from human macrophages, and our results ENSEMBL, LMP, and PDB as data sources and H. sapiens as an organism. demonstrate that calpain activity is required for NLRP3 inflamma- The identified proteins were compared with database ExoCarta version 4.1 some activation, as well as for the activation of unconventional containing all known exosomal proteins (30). Proteins were also analyzed with the prediction tool SignalP version 4.1 (31) to recognize the presence of vesicle-mediated protein secretion in macrophages in response to signal peptides required for conventional protein secretion. The functional ATP stimulation. annotation tool, the Database for Annotation, Visualization and Integrated Discovery (DAVID) (32) was used to find enriched pathways and networks Materials and Methods among secreted proteins. Cells and reagents Western blotting Human peripheral blood–derived monocytes were isolated from leukocyte- For Western blotting of whole-cell lysates, cells were collected in lysis buffer rich buffy coats from healthy blood donors (The Finnish Red Cross Blood (10 mM Tris [pH 7.4], 150 mM NaCl, 25% ethylene glycol) supplemented Transfusion Service) and differentiated into macrophages in Macrophage- with Complete Mini protease inhibitor (Roche Applied Science) and ho- SFM (Life Technologies) supplemented with 10 ng/ml GM-CSF (BioSource mogenized by ultrasound sonication. For Western blotting of cell fractions, International) and antibiotics for 6 d, as described (23). cells were collected and fractionated using a Qproteome Cell Compartment Bone marrow cells were isolated from femurs of male C57BL/6 wild- 2/2 Kit (QIAGEN), according to the manufacturer’s instructions. A total of 10 mg type mice (Scanbur) and male P2RX7 mice (The Jackson Labora- of protein of cell lysates, or a protein amount that was equal to 10% of cell tory). Bone marrow–derived macrophages (BMDMs) were differentiated culture supernatants, was loaded onto TGX gels (Bio-Rad) for SDS-PAGE by culturing bone marrow cells for 6 d in media containing 20% L929- and subsequently transferred to Immobilon-P Transfer Membranes (Merck conditioned media and 80% RPMI 1640 supplemented with 10% FBS, Millipore). Primary Abs were incubated overnight at 4˚C, and incubation L-glutamine, nonessential amino acids, 2-ME, sodium pyruvate, and anti- with appropriate HRP-conjugated secondary Abs (Dako) was done at room biotics. Cells were counted on day 7, and experiments were performed temperature for 1 h. Proteins were visualized with Western Lightning Plus on day 8. The experiments were performed in accordance with the ECL reagents (Perkin Elmer) using a charge coupled device camera European Convention for the Protection of Vertebrate Animals Used for (ImageQuant LAS 4000 Mini; GE Healthcare). The following Abs were Experimental and Other Scientific Purposes (Strasbourg March 18, used in the study: calpain-2, caspase-1, talin-1 (all from Sigma-Aldrich), 1986, adopted in Finland May 31, 1990). The study was approved by the Alix, calpain-1, galectin-3, GAPDH, LAMP-1, tsg 101 (all from Santa Animal Experiment Board and the State Provincial Office of Southern Cruz Biotechnology), a/b-tubulin, IL-1b (both from Cell Signaling), Finland. ASC (Millipore), and integrin a-X (ITGAX; CD11c; Novus Biologicals). The THP-1 human monocytic leukemia cell line was obtained from the American Type Culture Collection. The cells were activated with ultrapure Calpain activity assay Escherichia coli 0111:B4 LPS (1 mg/ml; InvivoGen) and recombinant human IFN-g (1000 U/ml; ImmunoTools) for 48 h before the experiments, Calpain activity from BMDMs was measured using a Calpain-Glo Protease as previously described (24). Assay (Promega) as described (33), with minor modifications. Briefly, ATP (3 mM), LPS (1 mg/ml, E. coli 0111:B4), curdlan (30 mg/ml), and BMDMs were grown as described above. Cells were plated on a white nigericin (10 mM) were purchased from Sigma-Aldrich. Monosodium 96-well plate and allowed to attach overnight. Calpain substrate (20 mM) The Journal of Immunology 3 was added to cells, and cells were incubated for 30 min. Then, cells were and functional studies. For secretome analysis, the secreted proteins stimulated with 3 mM ATP for 30 min, media were removed, and cells from unstimulated and ATP-activated macrophages were concen- m were lysed with 0.9% Triton-X (50 l/well). Calpain-Glo luciferase de- trated, separated by SDS-PAGE, and visualized with silver staining tection reagent (1:1 with lysis buffer) was added to the wells, and the plate was incubated for 15 min at room temperature with gentle shaking. (Fig. 1A). This showed that a 15-min stimulation of macrophages The luminescence was read with a VICTOR3 Multilabel Plate Reader with ATP induces a massive secretion of proteins. The proteins (PerkinElmer). Statistical testing was done using the t test. from silver-stained gels were identified using gel electrophoresis Isolation of secreted vesicles, nanoparticle tracking analysis, LC-MS/MS, as described (14, 26). The analysis was performed on and electron microscopy three independent biological replicates. In total, 260 and 405 proteins were identified from control and ATP-stimulated macrophage Human macrophages were treated with MDL 28170 (100 mM) and stim- supernatants, respectively (Fig. 1A, Supplemental Table I). From ulated with 3 mM ATP for 15 min. Subsequently, supernatants were collected, and extracellular vesicles (EVs) were isolated with ultracentrifugation, these, 94 and 188 proteins were identified in at least two biological as described (34). Isolated vesicles were suspended in PBS for nanoparticle replicates of control and ATP-stimulated supernatants, respectively. tracking analysis (NTA) or for electron microscopy. NTA and electron mi- From the reproducible identifications, 99 proteins were identified croscopy were performed as previously described (34). Alternatively vesicles only from ATP-stimulated cell secretomes. were suspended in SDS-PAGE buffer for silver staining and Western blotting. To gain insight into the protein-secretion mechanisms activated ELISA and cytotoxicity assay by ATP, we compared the identified proteins with the ExoCarta Secretion of IL-1b and IL-1a was analyzed using human IL-1b Eli-Pair database containing all known exosome-associated proteins. We ELISA (Diaclone; Tepnel Research Product & Services) and a human also used SignalP to predict the presence of conventionally secreted IL-1a ELISA MAX Deluxe Set (BioLegend), respectively. Cytotoxicity proteins. These analyses showed that a large majority of the Downloaded from was measured with a Cytotoxicity Detection Kit (LDH) (Roche). Statis- identified proteins are known as exosome-associated proteins tical testing was performing using the t test. All kits were used according (Fig. 1A), suggesting that ATP stimulation activates unconven- to the manufacturers’ instructions. tional vesicle-mediated protein secretion. To support this notion, ASC oligomerization assay weusedWesternblottingtoanalyze the secretion kinetics of Human primary macrophages were primed with 1 mg/ml LPS for 5.5 h and known exosomally secreted proteins (i.e., talin-1, ITGAX, then stimulated with 3 mM ATP for 30 min. The cells were treated with a/b-tubulin, and galectin-3) that were identified in the secretomes MDL 28170 (100 mM) for 1 h before ATP stimulation in indicated groups. of ATP-stimulated macrophages (Fig. 1B). This analysis clearly http://www.jimmunol.org/ Subsequently, an ASC oligomerization assay was performed, as described showed that they are already secreted from macrophages at 15 min (35, 36). Briefly, cells were rinsed with ice-cold PBS, collected in PBS, and pelleted by centrifugation. The cell pellet was suspended in lysis after ATP stimulation. Next, we enriched the fraction containing buffer (20 mM HEPES-KOH [pH 7.5], 150 mM KCl, 1% Nonidet P-40, EVs from the cell culture media by ultracentrifugation. Electron 0.1 mM PMSF, protease inhibitor) and homogenized by shearing with a microscopy analysis revealed large amounts of EVs with a char- 23g needle and syringe. The detergent-insoluble fraction was pelleted by acteristic exosomal size and shape after ATP stimulation (Fig. 1C). centrifugation at 5000 3 g at 4˚C for 10 min. Supernatant was collected as NTA showed that the majority of these secreted vesicles had sizes the soluble fraction. The pellet was suspended in PBS, and oligomerized ASC was cross-linked with 2 mM disuccinimidyl suberate (Sigma) for 30 min between 101 and 250 nm (Fig. 1D). To confirm that ATP-induced

at 37˚C, after which the insoluble fraction was pelleted by centrifugation at vesicle secretion is not due to apoptosis in human macrophages, we by guest on October 1, 2021 5000 3 g at 4˚C for 10 min and suspended in SDS buffer. ASC was detected analyzed caspase-3 activation, which is a hallmark of apoptosis, in with Western blotting from detergent-soluble and insoluble fractions. ATP-activated macrophages. ATP did not activate caspase-3 (data not Calcium imaging shown), demonstrating that apoptosis is not induced in ATP-treated macrophages. Taken together, our experiments show that robust un- Calcium imaging was performed as previously described (37). Human peripheral blood–derived monocytes were differentiated into macrophages on conventional vesicle-mediated protein secretion is triggered from cover glasses (VWR International) for 6 d. Then cells were incubated with human macrophages upon ATP stimulation. 4 mM Fura-2–acetoxymethyl ester (Molecular Probes Invitrogen; Life Technologies, Paisley, U.K.) dissolved in DMSO (Sigma-Aldrich) at 37˚C Calpains are activated in macrophages upon ATP stimulation for 20 min in HEPES-buffered media (pH 7.4) consisting of 137 mM NaCl, Next, we classified the identified secretome proteins based on their 5mMKCl,0.44mMKHPO , 4.2 mM NaHCO , 2 mM CaCl ,0.5mM 2 4 3 2 known cellular location. This showed that most of the proteins MgCl2, 10 mM HEPES, 10 mM glucose (all from Sigma), and 0.5 mM probenecid (to prevent leakage of Fura-2) before imaging. After incubation, secreted from ATP-stimulated macrophages were intracellular or cover glasses were attached to a temperated perfusion chamber in a Nikon membrane proteins (Fig. 2A), which is well in-line with the ob- TMS inverted microscope (203 objective) and perfused at 2 ml/min at 37˚C. servation that ATP activates unconventional vesicle-mediated Using a 340- and 380-nm light excitation filter changer under the control of protein secretion. In total, 91 membrane proteins were identified the InCytIM-2 System (Intracellular Imaging, Cincinnati, OH) and a dichroic mirror (DM430; Nikon), up to 100 cells could be recorded simultaneously. from ATP-stimulated cell secretomes compared with 34 mem- Light emission was measured through a 510-nm barrier filter with an in- brane proteins from unstimulated cell secretomes (Supplemental tegrating charge coupled device camera (Cohu, Poway, CA). A ratioed Table II). The 57 membrane proteins that were identified only (340/380 nm) image was acquired each second. The data collected were from the secretomes of ATP-stimulated cells were further ana- analyzed with InCyt 4.5 software (Intracellular Imaging) and further processed with Origin 6.0 software (OriginLab, Northampton, MA). lyzed using the functional annotation tool DAVID. GO: Biological Process, BioCarta, and Kyoto Encyclopedia of Genes and Ge- nomes pathway analyses all indicated that cell adhesion–related Results proteins as well as proteins related to regulation of actin cyto- ATP stimulation activates vesicle-mediated unconventional skeleton organization are enriched among secreted membrane protein secretion from human primary macrophages proteins. Consequently, our data strongly suggest that membrane- ATP-driven activation of P2X7 is associated with several membrane- trafficking events are activated in macrophages upon ATP stimulation. trafficking events (38); however, the total protein secretion and BioCarta pathway analysis showed that two calpain-related path- associated intracellular signaling pathways upon ATP activation of ways related to cell spread and motility were overrepresented in human macrophages are not well characterized. In this article, we our dataset (Fig. 2A). Calpains are cytosolic proteases that are acti- provide global secretome characterization of human primary mac- vated when intracellular Ca2+ concentration increases. It is well rophages stimulated with ATP using proteomics, bioinformatics, documented that activation of P2X7 causes alterations in cellular 4 CALPAINS AND ATP-DRIVEN PROTEIN SECRETION

FIGURE 1. Secretome analysis shows that ATP stimulation activates rapid vesicle-mediated unconventional protein secretion from human macrophages. (A) Human macrophages were left unstimulated or were stimulated with 3 mM ATP for 0.5 h. Cell culture supernatants were collected and concentrated, and proteins in the concentrated supernatants were separated by SDS-PAGE. Proteins were visualized by silver staining, followed by protein identification by gel electrophoresis LC-MS/MS. The secreted proteins were identified from three independent biological experiments, and the numbers of identified proteins are shown. The proteins identified in at least two replicate experiments were subjected to bioinformatics Downloaded from analysis. Proteins found in the database ExoCarta were designated as exosomal proteins. Proteins having a signal peptide identified by SignalP were designated as conventionally secreted proteins. (B)Westernblot analysis of secretion of talin-1, ITGAX, galectin-3, and a/b-tubulin from primary macrophages stimulated with 3 mM ATP for the indicated times. Representative re- http://www.jimmunol.org/ sults from two experiments are shown. (C) Vesicles were isolated from cell culture supernatants as described in Materials and Methods. Electron microscopy images of the vesicles were captured. Scale bars, 1 mm. (D)Size distribution of isolated vesicles was measured using NTA. The data obtained from two analyses are shown. by guest on October 1, 2021

ionic balance, including elevation of intracellular concentration upon ATP stimulation of P2X7-deﬁcient BMDMs (Fig. 2E). These of Ca2+. Therefore, we explored the role of calpains in ATP- results demonstrate that ATP-induced calpain activation is depen- induced protein secretion in human macrophages. First, we analyzed dent on P2X7. the expression of two isoforms of calpains with known protease activity in mammals, calpain-1 and calpain-2, in macrophages upon ATP Calpain activity is required for ATP-driven vesicle-mediated stimulation. Both calpains were detected from macrophages, and their protein secretion from macrophages amounts increased in whole-cell lysates during ATP stimulation Secretome analysis showed that extracellular ATP strongly acti- (Fig. 2B). Some activated calpains migrate to plasma membrane to vates vesicle-mediated protein secretion. We next elucidated the associate with phospholipids or protein complexes after which they are role of calpains in this process. First, we analyzed the secretion of released to the cytosol (11). We next analyzed the amounts of calpains talin-1, ITGAX, a/b-tubulin, and galectin-3 in ATP-activated from the cytosolic and membrane fractions of ATP-stimulated human macrophages that were left untreated or treated with calpain in- primary macrophages. Extracellular ATP increased the amount of hibitor MDL 28170. Strikingly, the secretion of these proteins was calpains in both cell fractions, implying that calpains are activated completely absent from MDL 28170-treated cells (Fig. 3A). during ATP stimulation (Fig. 2C). We used AZ 10606120, which is a Furthermore, calpeptin, which is a potent and cell-permeable selective antagonist of the receptor, to analyze whether ATP-induced calpain inhibitor that is chemically different from MDL 28170, calpain activation is mediated through P2X7 in human macrophages. also decreased secretion of talin-1, ITGAX, and a/b-tubulin in AZ 10606120 clearly inhibited the ATP-induced increase in calpain-1 ATP-activated macrophages (Supplemental Fig. 1). Next, we and -2 in the cytosolic fractions of human macrophages (Fig. 2D), isolated EVs from cell culture supernatants of unstimulated and demonstrating a direct role for P2X7 in calpain activation. ATP-stimulated cells with and without MDL 28170 treatment. The To directly determine the activation of calpains, we used a proteins from isolated EVs were separated by SDS-PAGE and luminescence-based assay to measure calpain activity in cell lysates visualized with silver staining. This showed that MDL 28170 of unstimulated and ATP-stimulated mouse BMDMs. Calpain ac- completely inhibited ATP-induced vesicular protein secretion tivity was clearly elevated in wild-type macrophages stimulated with (Fig. 3A). Analysis of exosomal marker proteins Alix and tsg 101 ATP. In contrast, calpain activity was comparable to control cells from these samples further conﬁrmed that inhibition of calpains The Journal of Immunology 5 Downloaded from http://www.jimmunol.org/

FIGURE 2. Calpains are activated upon ATP stimulation of macrophages, and activation is dependent on the P2X7 receptor. (A) Proteins identified in at least two replicate experiments were classified by GO class cellular component. The membrane proteins identified only from ATP-stimulated samples

(Supplemental Table II) were further analyzed by the functional annotation tool DAVID, and results of GO term enrichment reflecting the highest p values by guest on October 1, 2021 from BioCarta and Kyoto Encyclopedia of Genes and Genomes pathway analysis are shown. (B) Human macrophages were left unstimulated or were stimulated with 3 mM ATP for 0.5 h. Subsequently, cell lysates were collected, and the expression of calpain-1 (CANP-1) and calpain-2 (CANP-2) was analyzed by Western blotting. (C) Cells were stimulated as in (B). Cytosolic and membrane fractions of the cells were isolated, and the expression of calpain-1 and calpain-2 was analyzed by Western blotting. A representative result from two experiments is shown. (D) Human macrophages were left unstimulated or were stimulated with 3 mM ATP for 0.5 h in the absence and presence of the P2X7 receptor antagonist AZ 10606120 (AZ; 10 mM). Cytosolic fractions of the cells were isolated, and the expression of calpain-1 and calpain-2 was analyzed by Western blotting. (E) BMDMs from wild-type mice and P2RX72/2 mice were left untreated or were stimulated with 3 mM ATP for 0.5 h. Calpain activity in cell lysates was analyzed. Results (mean and SD) from four replicate experiments are shown. *p , 0.05. blocks vesicular protein secretion from human macrophages upon pathway (39). Therefore, we studied whether the secretion of this ATP stimulation (Fig. 3A). The secreted EVs were next studied by cytokine is dependent on calpain activity. Macrophages were first electron microscopy, which showed that the EV secretion induced primed with LPS to induce expression of pro–IL-1b protein. by ATP was completely absent when the cells were treated with After priming, macrophages were stimulated with ATP in the MDL 28170 (Fig. 3B). NTA showed that ATP-induced secretion absence and presence of increasing concentrations of MDL of EVs was 200-fold higher compared with control samples and 28170. MDL 28170 decreased the secretion of IL-1b in a dose- that inhibition of calpain activity with MDL 28170 completely dependent manner (Fig. 4A). MDL 28170 had no effect on the abolished the ATP-induced vesicle secretion (Fig. 3C). transcription of IL-1b (data not shown) or the level of pro–IL-1b Subsequently, we analyzed whether ATP-induced vesicle-mediated protein (Fig. 4B), demonstrating that MDL 28170 specifically protein secretion is dependent on P2X7. Human macrophages were inhibits the release of IL-1b but not the synthesis of pro–IL-1b. activatedwithATPinthepresenceandabsenceoftheselectiveP2X7 It was shown that processing and secretion of IL-1a, a proin- antagonist AZ 10606120, after which secretion of talin-1, ITGAX, flammatory IL-1 family cytokine, are calpain dependent (10, 12). a/b-tubulin, and galectin-3 was analyzed by Western blotting. AZ In accordance with this, MDL 28170 inhibited ATP-induced IL-1a 10606120 completely inhibited ATP-induced secretion of these pro- release in LPS-primed human macrophages in a dose-dependent teins (Fig. 3D), demonstrating that ATP-induced vesicle-mediated manner (Fig. 3C). protein secretion is dependent on P2X7. To confirm that calpain activity is required for the secretion of IL-1b, we used other calpain inhibitors (calpeptin and PD 150606) Calpain activity is required for ATP-driven secretion of IL-1 in addition to MDL 28170. Calpeptin (Fig. 4D) and PD 150606 family cytokines (Fig. 4E) inhibited the secretion of IL-1b from LPS-primed IL-1b is a proinflammatory cytokine that was suggested to be released macrophages stimulated with ATP. Elevation of the intracellular from macrophages through an unconventional protein-secretion concentration of Ca2+ is required for calpain activation; therefore, 6 CALPAINS AND ATP-DRIVEN PROTEIN SECRETION

FIGURE 3. Inhibition of calpain activity blocks vesicle-mediated protein secretion upon ATP stimulation. (A) Human macrophages were treated with MDL 28170 (100 mM) for 1 h, after which cells were left unstimulated or were stimulated with 3 mM ATP for 15 min. Cell culture supernatants were collected and concentrated, and the secretion of proteins talin-1, ITGAX, galectin-3, and a/b-tubulin was analyzed by Western blotting. Next, EVs were isolated from concentrated cell culture supernatants, and proteins in the isolated vesicles were separated by SDS-PAGE and visualized by silver staining. The presence of exosomal marker proteins Alix and tsg 101 in the vesicle fraction was determined by Western blotting. (B) Electron microscopy images of puriﬁed EVs. Scale bars, 1 mm. (C) NTA of EVs. The D data obtained from two analyses are shown. ( ) Downloaded from Human macrophages were left untreated or were treated with AZ 10606120 (AZ; 10 mM) for 1 h, after which cells were left unstimulated or were stimulated with 3 mM ATP for 15 min. Subsequently, cell culture supernatants were collected and concentrated, and the secretion of proteins talin-1, ITGAX, galectin-3, and http://www.jimmunol.org/ a/b-tubulin was analyzed by Western blotting.

we used the Ca2+ chelator BAPTA-AM to study the requirement activation, we primed human macrophages with LPS and stimu- of Ca2+ for IL-1b secretion. Like calpain inhibitors, BAPTA-AM lated them with ATP, with and without MDL 28170. We extracted reduced the amount of secreted IL-1b (Fig. 4F), further demon- the detergent-insoluble fraction from the cell lysates, and ASC 2+

strating that Ca -activated calpains are essential for IL-1b secre- oligomerization was analyzed by Western blotting (Fig. 6A). Di- by guest on October 1, 2021 tion in response to ATP stimulation. mers and oligomers of ASC were readily detected in samples The major cell wall components of fungi, 1,3-b–glucans and the prepared from cells stimulated with LPS and ATP, but MDL 28170 microbial toxin nigericin, are well-established inducers of IL-1b clearly inhibited ASC oligomerization similar to activated cells. secretion (6, 40). To assess how conserved the requirement of ASC dimers and oligomers were not observed in MDL 28170– calpain activity is for IL-1b release, we stimulated human mac- treated cells; however, a pronounced ASC aggregate (50–75 kDa) rophages with 1,3-b–glucans curdlan and with LPS and nigericin was detected. and studied the secretion of IL-1b in the absence and presence of Secretion of inflammasome components was detected upon MDL 28170. Interestingly, MDL 28170 blocked the curdlan- inflammasome activation (40, 43, 44). We next analyzed the se- induced secretion of IL-1b (Fig. 4G), whereas nigericin-induced cretion of ASC and the active form of caspase-1, p20, from LPS- IL-1b secretion was increased (Fig. 4H). primed and ATP-stimulated human macrophages. MDL 28170 treatment of macrophages completely abolished ATP-induced ATP-induced necrosis is dependent on calpain activity secretion of ASC and the active form of caspase-1 (Fig. 6B). In- Prolonged activation of cells with ATP leads to cell necrosis, which terestingly, ASC and the active form of caspase-1 were efficiently is associated with lactate dehydrogenase (LDH) release from the secreted from ATP-stimulated human macrophages in the absence cells (41). Therefore, we studied whether calpain activity is re- of LPS priming. These data indicate that inflammasome activation quired for necrosis triggered by ATP stimulation. We first mea- is independent of LPS priming in human primary macrophages. sured LDH release from macrophages upon prolonged ATP exposure. We also analyzed the secretion of calpain-1 and -2; similarly to LDH release increased in a time-dependent manner from ATP-treated inflammasome components, their secretion was abrogated by human macrophages (Fig. 5A). Strikingly, LDH release from ATP- MDL 28170 treatment. Similar results were observed in THP-1 activated cells was totally blocked when the human macrophages cells (Supplemental Fig. 2). were pretreated with MDL 28170 (Fig. 5B). This was verified in EVs released from human macrophages upon P2X7 activation THP-1 cells, even with extended ATP exposure up to 4 h (Fig. 5C). contain inflammasome components Calpains act upstream of the NLRP3 inflammasome Next, we studied whether inflammasome components and IL-1b Extracellular ATP is a well-known activator of the NLRP3 are secreted from ATP-stimulated human macrophages through a inflammasome, a protein complex that activates protease caspase-1 vesicle-mediated secretion pathway. LPS was used to induce the and leads to secretion of the cleaved biologically active form of production of pro–IL-1b protein; subsequently, cells were stimu- IL-1b. ASC is oligomerized upon inflammasome activation, and lated with ATP, supernatants were collected, and EVs were iso- the oligomers are detected as detergent-insoluble aggregates (42). To lated from cell culture media. Interestingly, Western blot analysis study whether calpain activity is essential for NLRP3 inflammasome showed that the adaptor protein ASC and activated caspase-1 The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 4. ATP-induced secretion of IL-1b and IL-1a is calpain dependent. (A) Human macrophages were primed with LPS (1 mg/ml) for 22 h. After priming, the cells were treated with the indicated concentrations of MDL 28170 for 1 h and stimulated with 3 mM ATP for 1 h. The cell culture supernatants were collected, and secretion of IL-1b was analyzed by ELISA. (B) Human macrophages were primed with LPS (1 mg/ml) for 4.5 h. Then cells were treated with MDL 28170 (100 mM) for 1 h, after which they were stimulated with 3 mM ATP for 0.5 h. Cell lysates were prepared, and pro–IL-1b expression was studied using Western blotting. (C) Human macrophages were primed with LPS (1 mg/ml) for 23 h. Subsequently, cells were treated with the indicated concentrations of MDL 28170 for 1 h and then stimulated with 3 mM ATP for 0.5 h. Cell culture supernatants of macrophages prepared from PBMCs from three bloods donors were collected and pooled together, and the secretion of IL-1a was analyzed with IL-1a ELISA. (D–F) Human macrophages were primed with LPS and stimulated with ATP, as in (B). One hour before stimulation with ATP, the cells were treated with calpeptin (100 mM), PD 150606 (100 mM), or BAPTA-AM (10 mM). Cell culture supernatants were collected, and IL-1b secretion was analyzed by ELISA. Representative results from three experiments are shown. (G and H) Cells were stimulated with curdlan (30 mg/ml) for 18 h in the absence and presence of MDL 28170 (100 mM), or they were primed with LPS (1 mg/ml) for 3 h, treated with MDL 28170 (100 mM) for 1 h, and stimulated with nigericin (10 mM) for 2 h. Cell culture supernatants were collected, and IL-1b secretion was analyzed by ELISA. A representative result from three experiments is shown. Data are mean 6 SD. were present in EVs secreted from LPS-primed macrophages Calpain inhibition does not block ATP-induced increase in stimulated with ATP (Fig. 6D). Activated caspase-1 also was de- intracellular Ca2+ tected in the EV fraction of unprimed ATP-stimulated macrophage ATP stimulation of macrophages is known to induce calcium inﬂux, secretomes. Unexpectedly, we did not detect the mature form of which is a proximal read-out of the P2X7 channel activation. Some IL-1b in the EV fraction; however, pro–IL-1b was present in EVs studies also suggested a requirement for calcium in assembly of the released by LPS-primed and ATP-activated macrophages. NLRP3 inﬂammasome (45, 46). Therefore, we next studied the 8 CALPAINS AND ATP-DRIVEN PROTEIN SECRETION

FIGURE 5. ATP-induced cell necrosis is dependent on calpain activity. (A) Macrophages were left untreated or were treated for the indicated times with 3 mM ATP. Cell culture supernatants were collected, and cytotoxicity was determined by measurement of LDH activity. Mean and SD from three donors are shown. (B) Human macrophages were left untreated or were treated with MDL 28170 (100 mM) for 1 h. Subsequently, the cells were stimulated with 3 mM ATP for the times indicated. Results from macrophages prepared from PBMCs from three blood donors are shown. (C) THP-1 cells were left untreated or were treated with MDL 28170 (100 mM) for 1 h before stimulation with 3 mM ATP for the indicated times. Cytotoxicity was measured as LDH activity from cell culture supernatants. The results from four experiments are shown. *p , 0.05, **p , 0.01. effect of MDL 28170 on intracellular Ca2+ concentrations upon had no effect on the overall protein secretion from macrophages ATP stimulation in Fura-2–loaded macrophages. In the absence of (Fig. 7A); however, as expected, caspase-1 inhibition decreased the MDL 28170, ATP induced an increase in intracellular Ca2+ con- secretion of mature IL-1b from LPS-primed ATP-stimulated mac- Downloaded from centrations within seconds; however, these changes were also rophages (data not shown). In line with the overall protein secretion clearly seen in cells treated with MDL 28170 (Fig. 6D). These not being affected by caspase-1 inhibition, secretion of the uncon- data indicate that MDL 28170 does not directly antagonize P2X7 ventionally released proteins talin-1, ITGAX, a/b-tubulin and galectin-3 channel activation. was not affected by Z-YVAD-FMK (Fig. 7B). We showed previously that cathepsin activity is essential for Caspase-1 activation is not required for unconventional unconventional protein secretion induced by MSU or b-glucans protein secretion upon ATP stimulation (14, 15). To study the possible involvement of cathepsins in ATP- http://www.jimmunol.org/ Activated caspase-1 was suggested to regulate unconventional induced vesicle-mediated protein secretion we used CA-074 Me, protein secretion (47). Therefore, we explored whether caspase-1 which is a speciﬁc inhibitor of cathepsins. CA-074 Me treatment activation is required for ATP-induced vesicle release. We left the of macrophages did not inhibit ATP-induced secretion of talin-1, macrophages untreated or treated them with the caspase-1 inhibitor ITGAX, a/b-tubulin, or galectin-3 (Fig. 7C); however, it inhibited Z-YVAD-FMK prior to ATP stimulation. Notably, Z-YVAD-FMK the vesicle-mediated secretion induced by MSU, demonstrating by guest on October 1, 2021

FIGURE 6. NLRP3 inflammasome activation is dependent on calpain activity. (A) Human macrophages were primed with LPS for 4.5 h, treated with MDL 28170 (100 mM) for 1 h, and stimulated with 3 mM ATP for 0.5 h. The detergent-soluble and insoluble fractions of the cell lysates were isolated, and ASC aggregates from the insoluble fraction were linked using disuccinimidyl suberate. The presence of ASC in fractions was analyzed using Western blotting. The experiment was repeated three times with similar results. (B) Human macrophages were primed with LPS (1 mg/ml) for 4.5 h. Then cells were treated with MDL 28170 (100 mM) for 1 h and stimulated with 3 mM ATP for 0.5 h. Cell culture supernatants were collected and concentrated, and the secretion of inflammasome-related proteins caspase-1 and ASC, as well as calpain-1 and calpain-2, was analyzed by Western blotting. (C) Human macrophages were primed with LPS (1 mg/ml) for 5.5 h and then cells were stimulated with 3 mM ATP for 0.5 h. Supernatants were collected and concentrated, and EVs were purified. The presence of pro–IL-1b, ASC, and caspase-1 in EVs was analyzed by Western blotting. A representative result from three experiments is shown. (D) Fura-2–loaded macrophages were activated with 1 mM ATP (left panel) or with 1 mM ATP in the presence of MDL 28170 (100 mM) (middle panel), or they were treated with MDL 28170 (100 mM) for 1 h and then activated with ATP (right panel). The increase in intracellular calcium was recorded using a fluorescence imaging system, as described in Materials and Methods. Each graph represents an average 6 SE of all individual imaged cells in the field (26–32) on each coverslip. A representative experiment from at least three experiments is shown. The Journal of Immunology 9

ventional protein secretion from human macrophages. Different types of EVs have been described; the main types are exosomes derived from multivesicular bodies, microvesicles shed from the plasma membrane, and apoptotic bodies (48). In this study, the majority of proteins secreted from ATP-stimulated cells were found in ExoCarta, a database of exosomal proteins derived from the literature (30). However, with current methods, the isolated preparation of EVs contains a heterogeneous mixture of vesicles (49, 50), and previous studies that relied on a few unique marker proteins associated with a particular vesicle type are being chal- lenged (51). Release of exosomes (22, 52) and microvesicle shedding (21, 53, 54) were suggested to be activated during ATP-driven triggering of P2X7. Our results clearly demonstrate that ATP stimulation activates release of exosomes and shedding of microvesicles from human macrophages in a P2X7-dependent manner. The analysis of secretion of membrane-associated proteins from ATP-stimulated macrophages revealed that signaling pathways associated with actin cytoskeleton organization, cell adhesion, and integrin signaling of the cells were enriched. Importantly, calpain signaling was among the enriched pathways. In accordance with Downloaded from this, we show that calpains are activated in macrophages upon ATP exposure. Previously, it was shown that calpain activity is elevated in mouse dendritic cells upon their stimulation by several inflammasome activators, including ATP (55). Calpains are Ca2+-activated nonlysosomal cysteine proteases (10), and intracellular Ca2+ concentrations are known to be elevated after activation of the P2X7 receptor http://www.jimmunol.org/ 2+ FIGURE 7. Caspase-1 or cathepsin activity is not required for uncon- (56). ATP-driven activation of P2X7 results in Ca influx. In our study, 2+ ventional protein secretion upon ATP stimulation. (A) Human macro- inhibition of calpains did not affect Ca influx after ATP stimulation, 2+ phages were left untreated or were treated with Z-YVAD-FMK for 1 h to demonstrating that calpains act downstream of Ca signaling. This inhibit caspase-1 activity, after which they were activated with 3 mM ATP provides a molecular explanation for the calpain activation seen in for 0.5 h. Subsequently, cell culture supernatants were collected and ATP-stimulated macrophages. Interestingly, in previous studies, release concentrated, and proteins were separated by SDS-PAGE and visualized of EVs and microvesicle shedding were blocked by inhibiting calcium B with silver staining. ( ) Human macrophages were left untreated or were influx via removal of extracellular Ca2+ (21, 22). Our results show that treated with Z-YVAD-FMK for 1 h, after which they were activated with ATP-induced secretion of EVs from macrophages is totally dependent 3 mM ATP for 0.5 h. Cell culture supernatants were collected and con- by guest on October 1, 2021 on calpain activity. centrated, and secretion of talin-1, ITGAX, galectin-3, and a/b-tubulin was analyzed by Western blotting. (C) Human macrophages were treated P2X7 on plasma membrane forms a multimeric signaling with the cathepsin-specific inhibitor CA-074 Me (25 mM) for 1 h, after complex composed of P2X7 receptors associated with other pro- which they were left unstimulated or were stimulated with 3 mM ATP for teins. The P2X7 complex was characterized in THP-1 cells and 30 min. Subsequently, cell culture supernatants were collected and con- P2X7-transfected HEK cells, as well as in cervical endothelial cells centrated, and the secretion of talin-1, ITGAX, galectin-3, and a/b-tubulin (57, 58); however, it has not been characterized in human primary was analyzed by Western blotting. (D) Human macrophages were treated cells. Cytoskeleton proteins (e.g., myosin, tubulin, and b-actin), with MSU (100 mg/ml) for 6 h in the absence and presence of CA-074 Me integrins, and different heat shock proteins are detected in com- (25 mM). Cell culture supernatants were collected and concentrated, and plex with P2X7. This suggests that calpain activation is a very vesicle-mediated protein secretion was analyzed by Western blotting. early response to ATP stimulation. The majority of known calpain substrates are cytoskeletal proteins and signaling proteins. We the functionality of the inhibitor (Fig. 6D). In conclusion, our speculate that, in macrophages, calpain cleavage of a cytoskeletal results show that calpains, but not caspase-1 or cathepsin activity, protein is required for P2X7 signaling downstream of calcium are required for activation of unconventional protein secretion in influx. Interestingly, Kim et al. (57) found that, in THP-1 cells, human macrophages in response to ATP stimulation. P2X7 is in complex with integrin b2 and b-actin, both of which are calpain substrates according to the calpain substrate database Discussion CaMPDB (59). However, additional studies are needed to identify Macrophages are the central effector cells of the innate immune calpain substrates that are essential for P2X7 signaling. system. When macrophages are activated, they signal to their In this study, we showed that calpain activity was required for the environment by secreting proteins that induce inflammatory re- secretion of the mature form of IL-1b upon ATP stimulation. The sponses, recruit other cells to the site of inflammation and con- secretion of IL-1b induced by ATP stimulation is regulated by tribute to tissue regeneration. Characterization of the secretome, the NLRP3 inflammasome–mediated activation of caspase-1 (6). Cleaved global protein secretion, from macrophages provides new insights IL-1b is rapidly secreted by the cells to initiate inflammatory re- into early innate immune responses. In this study, we activated sponses by binding to IL-1Rs on other cells. Previously, it was shown human macrophages with extracellular ATP, a well-acknowledged that calcium signaling is important for ATP-induced secretion of danger signal for innate immunity, to provide in-depth charac- IL-1b (55, 60, 61). Our findings that the activation of calpains is terization of the macrophage secretome upon ATP activation. required for secretion of IL-1b provides a possible explanation for We show that stimulation of human macrophages with ATP previous observations. Interestingly, although blocking of calpain induces robust and rapid (within minutes) protein secretion. More activity abrogated the secretion of IL-1b upon 1,3-b-glucans stimu- specifically, we show that ATP induces vesicle-mediated uncon- lation, it did not have an effect on nigericin-induced IL-1b secretion. 10 CALPAINS AND ATP-DRIVEN PROTEIN SECRETION

Nigericin is a potassium ionophore that permeabilizes the plasma appears in extracellular media in 15 min (22). In conclusion, these membrane and induces K+ efflux independently of P2X7 activation results and our findings indicate that ATP stimulation of macro- (62, 63). Differences in the signal cascades used by ATP and nigericin phages results in rapid activation of unconventional protein secre- might explain why calpains are dispensable for IL-1b secretion tion, which is followed by necrosis of macrophages and associated upon nigericin exposure; however, our results suggest that ATP secretion of IL-1b. and b-glucan–activated Dectin-1 pathways share the same sig- In conclusion, our results demonstrate that calpain activity is naling components, leading to calpain activation, following se- required for two independent signaling events in human macro- cretion of IL-1b. phages in response to ATP stimulation. First, calpains are required Previous studies showed that K+ efflux is essential for the for NLRP3 inflammasome activation, following necrosis and function of the NLRP3 inflammasome with all known activators along with associated IL-1b secretion. Second, activation of un- (44). Recently, it was found that calcium signaling is required for conventional vesicle-mediated protein secretion was dependent secretion events, as well as for the assembly and activation of the on calpain activity but was independent of cathepsins and NLRP3 inflammasome (45, 46). However, it also was suggested inflammasome-associated caspase-1. Taken together, our results that the role of Ca2+ in NLRP3 inflammasome activation may be demonstrate a pivotal role for calpains in the activation of the indirect or cell-model specific (64, 65). In this study, we demon- inflammatory response during ATP exposure and suggest that strated that the NLRP3 inflammasome complex was not assembled small molecule inhibitors of calpains are potential therapeutics for normally when calpain activity was blocked. Furthermore, Ca2+ the treatment of inflammatory diseases. influx was not affected by inhibition of calpain activity. In conclusion, our results demonstrate that calpains act upstream of

Acknowledgments Downloaded from 2+ NLRP3 inflammasome activation but downstream of Ca signaling We thank Dr. Ville Veckman for assistance with BMDMs and Dr. Pia in ATP-activated macrophages. Siljander and Maria Aatonen (Department of Biochemistry, University of We also showed that, upon ATP stimulation, active caspase-1 and Helsinki) for help with NTA analysis. ASC were secreted from LPS-primed and unprimed human primary macrophages. This implies that, in human macrophages, the Disclosures priming step is needed only for the production of pro–IL-1b, and The authors have no financial conflicts of interest. there is a sufficient amount of NLRP3 protein to induce the as- http://www.jimmunol.org/ sembly of the inflammasome when an activating signal is detected. This is in contrast to mouse macrophages, in which priming is References necessary for the production of NLRP3 and pro–IL-1b (66). In 1. Janeway Jr., C. A. 1989. Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harb. Symp. Quant. Biol. 54: 1–13. this study, we also showed that, in human macrophages, ATP 2. Matzinger, P. 1994. Tolerance, danger, and the extended family. Annu. Rev. stimulation alone is sufficient to activate a robust vesicle-mediated Immunol. 12: 991–1045. 3. Iwasaki, A., and R. Medzhitov. 2010. Regulation of adaptive immunity by the protein secretion. This is in contrast to a study using murine innate immune system. Science 327: 291–295. macrophages, in which LPS priming was required to activate the 4. Solle, M., J. Labasi, D. G. Perregaux, E. Stam, N. Petrushova, B. H. Koller, secretion of lysosomes upon ATP exposure (22). We also observed R. J. Griffiths, and C. A. Gabel. 2001. Altered cytokine production in mice lacking P2X(7) receptors. J. Biol. Chem. 276: 125–132. by guest on October 1, 2021 in our studies that murine macrophages require a priming step to 5. Bartlett, R., L. Stokes, and R. Sluyter. 2014. The P2X7 receptor channel: recent activate robust protein secretion upon ATP stimulation (E. Va¨lima¨ki, developments and the use of P2X7 antagonists in models of disease. Pharmacol. W. Cypryk, T. Nyman, and S. Matikainen, unpublished observa- Rev. 66: 638–675. 6. Mariathasan, S., D. S. Weiss, K. Newton, J. McBride, K. O’Rourke, M. Roose- tions). This highlights the differences in signaling pathways between Girma, W. P. Lee, Y. Weinrauch, D. M. Monack, and V. M. Dixit. 2006. Cryopyrin human and murine macrophages and emphasizes that data derived activates the inflammasome in response to toxins and ATP. Nature 440: 228–232. only from murine models should be extrapolated to humans using 7. Martinon, F., K. Burns, and J. Tschopp. 2002. The inflammasome: a molecular platform triggering activation of inflammatory caspases and processing of proIL- caution. b. Mol. Cell 10: 417–426. Assembly of the NLRP3 inflammasome leads to activation of 8. Agostini, L., F. Martinon, K. Burns, M. F. McDermott, P. N. Hawkins, and caspase-1, which was suggested to regulate global unconventional J. Tschopp. 2004. NALP3 forms an IL-1b-processing inflammasome with increased activity in Muckle-Wells autoinflammatory disorder. Immunity 20: 319–325. protein secretion induced by UV exposure of human fibroblasts 9. Ono, Y., and H. Sorimachi. 2012. Calpains: an elaborate proteolytic system. (47). We showed that the NLRP3 inflammasome activators MSU Biochim. Biophys. Acta 1824: 224–236. 10. Goll, D. E., V. F. Thompson, H. Li, W. Wei, and J. Cong. 2003. The calpain (14), 1,3-b-glucans (15), and influenza A virus infection (17) in- system. Physiol. Rev. 83: 731–801. duce a robust protein secretion from human primary macrophages 11. Sato, K., and S. Kawashima. 2001. Calpain function in the modulation of signal along with activation of the NLRP3 inflammasome. In this study, transduction molecules. Biol. Chem. 382: 743–751. 12. Carruth, L. M., S. Demczuk, and S. B. Mizel. 1991. Involvement of a calpain- we showed that calpain activity is required for NLRP3 inflamma- like protease in the processing of the murine interleukin 1 alpha precursor. some activation and overall protein secretion from ATP-stimulated J. Biol. Chem. 266: 12162–12167. macrophages. It is not known whether the overall protein secretion 13. Gordon, S., A. Pluddemann,€ and F. Martinez Estrada. 2014. Macrophage hetero- geneity in tissues: phenotypic diversity and functions. Immunol. Rev. 262: 36–55. is dependent on inflammasome activation or whether they represent 14. Va¨lima¨ki, E., J. J. Miettinen, N. Lietzeń, S. Matikainen, and T. A. Nyman. 2013. parallel effects triggered by the same stimuli. To address this Monosodium urate activates Src/Pyk2/PI3 kinase and cathepsin dependent un- question, we treated macrophages with the caspase-1 inhibitor conventional protein secretion from human primary macrophages. Mol. Cell. Proteomics 12: 749–763. Z-YVAD-FMK before stimulation with ATP. Strikingly, overall 15. O¨ hman, T., L. Teirila¨, A.-M. Lahesmaa-Korpinen, W. Cypryk, V. Veckman, ATP-induced protein secretion was unaffected in cells treated with S. Saijo, H. Wolff, S. Hautaniemi, T. A. Nyman, and S. Matikainen. 2014. the caspase-1 inhibitor. It was shown very recently that diverse Dectin-1 pathway activates robust autophagy-dependent unconventional protein secretion in human macrophages. J. Immunol. 192: 5952–5962. activators of the NLRP3 inflammasome, including ATP, promote 16. Miettinen, J. J., S. Matikainen, and T. A. Nyman. 2012. Global secretome mature IL-1b secretion in THP-1 and J774 macrophage cell lines characterization of herpes simplex virus 1-infected human primary macrophages. J. Virol. 86: 12770–12778. by triggering necrosis (67). Similarly, our results show that ATP- 17. Lietzeń, N., T. O¨ hman, J. Rintahaka, I. Julkunen, T. Aittokallio, S. Matikainen, driven IL-1b secretion is associated with necrosis of human mac- and T. A. Nyman. 2011. Quantitative subcellular proteome and secretome pro- rophages. A previous report showed that plasma membrane filing of influenza A virus-infected human primary macrophages. PLoS Pathog. 7: e1001340. microvesicle shedding from LPS-primed ATP-stimulated murine 18. Nickel, W., and C. Rabouille. 2009. Mechanisms of regulated unconventional macrophages occurs 5 min after activation, whereas mature IL-1b protein secretion. Nat. Rev. Mol. Cell Biol. 10: 148–155. The Journal of Immunology 11

19. Raposo, G., and W. Stoorvogel. 2013. Extracellular vesicles: exosomes, 45. Murakami, T., J. Ockinger, J. Yu, V. Byles, A. McColl, A. M. Hofer, and microvesicles, and friends. J. Cell Biol. 200: 373–383. T. Horng. 2012. Critical role for calcium mobilization in activation of the 20. Virginio, C., A. MacKenzie, R. A. North, and A. Surprenant. 1999. Kinetics of NLRP3 inflammasome. Proc. Natl. Acad. Sci. USA 109: 11282–11287. cell lysis, dye uptake and permeability changes in cells expressing the rat P2X7 46. Lee, G.-S., N. Subramanian, A. I. Kim, I. Aksentijevich, R. Goldbach-Mansky, receptor. J. Physiol. 519: 335–346. D. B. Sacks, R. N. Germain, D. L. Kastner, and J. J. Chae. 2012. The calcium- 21. MacKenzie, A., H. L. Wilson, E. Kiss-Toth, S. K. Dower, R. A. North, and sensing receptor regulates the NLRP3 inflammasome through Ca2+ and cAMP. A. Surprenant. 2001. Rapid secretion of interleukin-1b by microvesicle shed- Nature 492: 123–127. ding. Immunity 15: 825–835. 47. Keller, M., A. Ruegg,€ S. Werner, and H.-D. Beer. 2008. Active caspase-1 is a 22. Qu, Y., L. Franchi, G. Nunez, and G. R. Dubyak. 2007. Nonclassical IL-1b regulator of unconventional protein secretion. Cell 132: 818–831. secretion stimulated by P2X7 receptors is dependent on inflammasome activa- 48. Gyo¨rgy, B., T. G. Szabo´,M.Pa´sztoí, Z. Pa´l, P. Misja´k, B. Aradi, V. La´szlo´, tion and correlated with exosome release in murine macrophages. J. Immunol. E. Pa´llinger, E. Pap, A. Kittel, et al. 2011. Membrane vesicles, current state-of-the- 179: 1913–1925. art: emerging role of extracellular vesicles. Cell. Mol. Life Sci. 68: 2667–2688. 23. Pirhonen, J., T. Sareneva, M. Kurimoto, I. Julkunen, and S. Matikainen. 1999. 49. van der Pol, E., A. N. Boïng, P. Harrison, A. Sturk, and R. Nieuwland. 2012. Virus infection activates IL-1 b and IL-18 production in human macrophages by Classification, functions, and clinical relevance of extracellular vesicles. Phar- a caspase-1–dependent pathway. J. Immunol. 162: 7322–7329. macol. Rev. 64: 676–705. 24. Humphreys, B. D., and G. R. Dubyak. 1998. Modulation of P2X7 nucleotide 50. Witwer, K. W., E. I. Buza´s, L. T. Bemis, A. Bora, C. Lasser, J. Lotvall, receptor expression by pro- and anti-inflammatory stimuli in THP-1 monocytes. ¨ ¨ J. Leukoc. Biol. 64: 265–273. E. N. Nolte-’t Hoen, M. G. Piper, S. Sivaraman, J. Skog, et al. 2013. Stan- 25. O’Connell, K. L., and J. T. Stults. 1997. Identification of mouse liver proteins on dardization of sample collection, isolation and analysis methods in extracellular two-dimensional electrophoresis gels by matrix-assisted laser desorption/ionization vesicle research. J. Extracell. Vesicles 2: 20360. mass spectrometry of in situ enzymatic digests. Electrophoresis 18: 349–359. 51. Bobrie, A., M. Colombo, S. Krumeich, G. Raposo, and C. The´ry. 2012. Diverse 26. Savijoki, K., N. Lietzeń, M. Kankainen, T. Alatossava, K. Koskenniemi, subpopulations of vesicles secreted by different intracellular mechanisms are P. Varmanen, and T. A. Nyman. 2011. Comparative proteome cataloging of present in exosome preparations obtained by differential ultracentrifugation. Lactobacillus rhamnosus strains GG and Lc705. J. Proteome Res. 10: 3460–3473. J. Extracell. Vesicles 1: 18397. 27. Elias, J. E., and S. P. Gygi. 2007. Target-decoy search strategy for increased 52. Qu, Y., L. Ramachandra, S. Mohr, L. Franchi, C. V. Harding, G. Nunez, and confidence in large-scale protein identifications by mass spectrometry. Nat. G. R. Dubyak. 2009. P2X7 receptor-stimulated secretion of MHC class Methods 4: 207–214. II-containing exosomes requires the ASC/NLRP3 inflammasome but is inde- Downloaded from 28. The Gene Ontology Consortium. 2000. Gene ontology: tool for the unification of pendent of caspase-1. J. Immunol. 182: 5052–5062. biology. Nat. Genet. 25: 25–29. 53. Pizzirani, C., D. Ferrari, P. Chiozzi, E. Adinolfi, D. Sandona`, E. Savaglio, and 29. Zeeberg, B. R., W. Feng, G. Wang, M. D. Wang, A. T. Fojo, M. Sunshine, F. Di Virgilio. 2007. Stimulation of P2 receptors causes release of IL-1b-loaded S. Narasimhan, D. W. Kane, W. C. Reinhold, S. Lababidi, et al. 2003. GoMiner: microvesicles from human dendritic cells. Blood 109: 3856–3864. a resource for biological interpretation of genomic and proteomic data. Genome 54. Bianco, F., E. Pravettoni, A. Colombo, U. Schenk, T. Mo¨ller, M. Matteoli, and Biol. 4: R28. C. Verderio. 2005. Astrocyte-derived ATP induces vesicle shedding and IL-1 b 30. Mathivanan, S., C. J. Fahner, G. E. Reid, and R. J. Simpson. 2012. ExoCarta release from microglia. J. Immunol. 174: 7268–7277.

2012: database of exosomal proteins, RNA and lipids. Nucleic Acids Res. 40: 55. Gross, O., A. S. Yazdi, C. J. Thomas, M. Masin, L. X. Heinz, G. Guarda, http://www.jimmunol.org/ D1241–D1244. M. Quadroni, S. K. Drexler, and J. Tschopp. 2012. Inflammasome activators 31. Petersen, T. N., S. Brunak, G. von Heijne, and H. Nielsen. 2011. SignalP 4.0: induce interleukin-1a secretion via distinct pathways with differential require- discriminating signal peptides from transmembrane regions. Nat. Methods 8: ment for the protease function of caspase-1. Immunity 36: 388–400. 785–786. 56. Wiley, J. S., R. Sluyter, B. J. Gu, L. Stokes, and S. J. Fuller. 2011. The human 32. Huang, W., B. T. Sherman, and R. A. Lempicki. 2009. Systematic and integrative P2X7 receptor and its role in innate immunity. Tissue Antigens 78: 321–332. analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc. 57. Kim, M., L. H. Jiang, H. L. Wilson, R. A. North, and A. Surprenant. 2001. 4: 44–57. Proteomic and functional evidence for a P2X7 receptor signalling complex. 33. Seyb, K. I., E. R. Schuman, J. Ni, M. M. Huang, M. L. Michaelis, and EMBO J. 20: 6347–6358. M. A. Glicksman. 2008. Identification of small molecule inhibitors of b-amyloid 58. Gu, B. J., C. Rathsam, L. Stokes, A. B. McGeachie, and J. S. Wiley. 2009. cytotoxicity through a cell-based high-throughput screening platform. J. Biomol. Screen. 13: 870–878. Extracellular ATP dissociates nonmuscle myosin from P2X(7) complex: this dissociation regulates P2X(7) pore formation. Am. J. Physiol. Cell Physiol. 297: 34. Cypryk, W., T. O¨ hman, E.-L. Eskelinen, S. Matikainen, and T. A. Nyman. 2014. by guest on October 1, 2021 Quantitative proteomics of extracellular vesicles released from human C430–C439. monocyte-derived macrophages upon b-glucan stimulation. J. Proteome Res. 13: 59. duVerle, D., I. Takigawa, Y. Ono, H. Sorimachi, and H. Mamitsuka. 2010. 2468–2477. CaMPDB: a resource for calpain and modulatory proteolysis. Genome Inform 35. Fernandes-Alnemri, T., J.-W. Yu, C. Juliana, L. Solorzano, S. Kang, J. Wu, 22: 202–213. P. Datta, M. McCormick, L. Huang, E. McDermott, et al. 2010. The AIM2 60. Brough, D., R. A. Le Feuvre, R. D. Wheeler, N. Solovyova, S. Hilfiker, inflammasome is critical for innate immunity to Francisella tularensis. Nat. N. J. Rothwell, and A. Verkhratsky. 2003. Ca2+ stores and Ca2+ entry differ- Immunol. 11: 385–393. entially contribute to the release of IL-1 b and IL-1 a from murine macrophages. 36. Zhao, C., D. D. Gillette, X. Li, Z. Zhang, and H. Wen. 2014. Nuclear factor E2- J. Immunol. 170: 3029–3036. related factor-2 (Nrf2) is required for NLRP3 and AIM2 inflammasome acti- 61. Suttles, J., J. G. Giri, and S. B. Mizel. 1990. IL-1 secretion by macrophages. vation. J. Biol. Chem. 289: 17020–17029. Enhancement of IL-1 secretion and processing by calcium ionophores. 37. Louhivuori, L. M., L. Jansson, P. M. Turunen, M. H. Ja¨ntti, T. Nordstro¨m, J. Immunol. 144: 175–182. V. Louhivuori, and K. E. A˚ kerman. 2015. Transient receptor potential channels 62. Perregaux, D., and C. A. Gabel. 1994. Interleukin-1 beta maturation and release and their role in modulating radial glial-neuronal interaction: a signaling path- in response to ATP and nigericin. Evidence that potassium depletion mediated by way involving mGluR5. Stem Cells Dev. 24: 701–713. these agents is a necessary and common feature of their activity. J. Biol. Chem. 38. Qu, Y., and G. R. Dubyak. 2009. P2X7 receptors regulate multiple types of 269: 15195–15203. membrane trafficking responses and non-classical secretion pathways. Purinergic 63. Le Feuvre, R. A., D. Brough, Y. Iwakura, K. Takeda, and N. J. Rothwell. 2002. Signal. 5: 163–173. Priming of macrophages with lipopolysaccharide potentiates P2X7-mediated 39. Dupont, N., S. Jiang, M. Pilli, W. Ornatowski, D. Bhattacharya, and V. Deretic. cell death via a caspase-1-dependent mechanism, independently of cytokine 2011. Autophagy-based unconventional secretory pathway for extracellular de- production. J. Biol. Chem. 277: 3210–3218. livery of IL-1b. EMBO J. 30: 4701–4711. 64. Katsnelson, M. A., L. G. Rucker, H. M. Russo, and G. R. Dubyak. 2015. K+ 40. Kankkunen, P., L. Teirila, J. Rintahaka, H. Alenius, H. Wolff, and S. Matikainen. ¨ efflux agonists induce NLRP3 inflammasome activation independently of Ca2+ 2010. (1,3)-b-glucans activate both dectin-1 and NLRP3 inflammasome in hu- signaling. J. Immunol. 194: 3937–3952. man macrophages. J. Immunol. 184: 6335–6342. 65. Vaeth, M., I. Zee, A. R. Concepcion, M. Maus, P. Shaw, C. Portal-Celhay, 41. Chan, F. K.-M., K. Moriwaki, and M. J. De Rosa. 2013. Detection of necrosis by release of lactate dehydrogenase activity. Methods Mol. Biol. 979: 65–70. A. Zahra, L. Kozhaya, C. Weidinger, J. Philips, et al. 2015. Ca2+ signaling but 42. Bryan, N. B., A. Dorfleutner, Y. Rojanasakul, and C. Stehlik. 2009. Activation of not store-operated Ca2+ entry is required for the function of macrophages and inflammasomes requires intracellular redistribution of the apoptotic speck-like dendritic cells. J. Immunol. 195: 1202–1217. protein containing a caspase recruitment domain. J. Immunol. 182: 3173–3182. 66. Bauernfeind, F. G., G. Horvath, A. Stutz, E. S. Alnemri, K. MacDonald, 43. Hornung, V., F. Bauernfeind, A. Halle, E. O. Samstad, H. Kono, K. L. Rock, D. Speert, T. Fernandes-Alnemri, J. Wu, B. G. Monks, K. A. Fitzgerald, et al. K. A. Fitzgerald, and E. Latz. 2008. Silica crystals and aluminum salts activate 2009. Cutting edge: NF-kappaB activating pattern recognition and cytokine re- the NALP3 inflammasome through phagosomal destabilization. Nat. Immunol. 9: ceptors license NLRP3 inflammasome activation by regulating NLRP3 expres- 847–856. sion. J. Immunol. 183: 787–791. 44. Martinon, F., V. Pe´trilli, A. Mayor, A. Tardivel, and J. Tschopp. 2006. Gout- 67. Cullen, S. P., C. J. Kearney, D. M. Clancy, and S. J. Martin. 2015. Diverse ac- associated uric acid crystals activate the NALP3 inflammasome. Nature 440: tivators of the NLRP3 inflammasome promote IL-1b secretion by triggering 237–241. necrosis. Cell Reports 11: 1535–1548.