Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Proteomic Analysis Reveals Distinct Metabolic Differences Between Granulocyte-Macrophage Colony Stimulating Factor (GM-CSF) and Macrophage Colony Stimulating Factor (M-CSF) Grown Macrophages Derived from Murine Bone Marrow Cells*

Proteomic Analysis Reveals Distinct Metabolic Differences Between Granulocyte-Macrophage Colony... crossmark Research © 2015 by The American Society for Biochemistry and Molecular Biology, Inc. This paper is available on line at http://www.mcponline.org Proteomic Analysis Reveals Distinct Metabolic Differences Between Granulocyte-Macrophage Colony Stimulating Factor (GM-CSF) and Macrophage Colony Stimulating Factor (M-CSF) Grown Macrophages Derived from □ S Murine Bone Marrow Cells* Yi Rang Na‡**, Ji Hye Hong§**, Min Yong Lee§, Jae Hun Jung§, Daun Jung‡, Kwang Pyo Kim§, Young Won Kim‡, Dain Son‡, Murim Choi¶, and Seung Hyeok Seok II‡ Macrophages are crucial in controlling infectious agents ulation in GM-BMMs depends on their acute glycolytic and tissue homeostasis. Macrophages require a wide capacity. In contrast, M-BMMs up-regulate proteins in- range of functional capabilities in order to fulfill distinct volved in endocytosis, which correlates with a tendency roles in our body, one being rapid and robust immune toward homeostatic functions such as scavenging cellu- responses. To gain insight into macrophage plasticity and lar debris. Together, our data describes a proteomic the key regulatory protein networks governing their spe- network that underlies the pro-inflammatory actions of cific functions, we performed quantitative analyses of the GM-BMMs as well as the homeostatic functions of proteome and phosphoproteome of murine primary GM- M-BMMs. Molecular & Cellular Proteomics 14: 10.1074/ CSF and M-CSF grown bone marrow derived macro- mcp.M115.048744, 2722–2732, 2015. phages (GM-BMMs and M-BMMs, respectively) using the latest isobaric tag based tandem mass tag (TMT) labeling and liquid chromatography-tandem mass spectrometry Macrophages are a heterogeneous population of immune (LC-MS/MS). Strikingly, metabolic processes emerged as cells that are essential for the initiation and resolution of a major difference between these macrophages. Specifi- pathogen- or tissue damage-induced inflammation (1). They cally, GM-BMMs show significant enrichment of proteins show remarkable plasticity that allows them to respond effi- involving glycolysis, the mevalonate pathway, and nitro- ciently to environmental signals and change their phenotype gen compound biosynthesis. This evidence of enhanced and physiology upon cytokine and microbial signaling (2). glycolytic capability in GM-BMMs is particularly signifi- These changes can give rise to populations of cells with cant regarding their pro-inflammatory responses, be- distinct functions that are phenotypically characterized by the cause increased production of cytokines upon LPS stim- production of pro-inflammatory and anti-inflammatory cyto- kines (3). Among the growth factors that affect macrophage activation states, two cytokines that appear to be important in From the ‡Department of Microbiology and Immunology, and In- stitute of Endemic Disease, Seoul National University College of controlling the functions of macrophage lineage populations Medicine, 103 Daehak-ro, Chongno-gu, Seoul 110-799, South Korea; in inflammatory conditions are granulocyte-macrophage col- §Department of Applied Chemistry, Kyung Hee University, 1732 Deo- ony stimulating factor (GM-CSF) and macrophage colony gyeong-daero, Giheung-gu, Yongin-si, Gyeonggi-do 446-701, South stimulating factor (M-CSF) (4). These CSFs are critical to the Korea; ¶Department of Biomedical Science, Seoul National University proper maintenance of steady-state macrophage develop- College of Medicine, 103 Daehak-ro, Chongno-gu, Seoul 110-799, South Korea ment, although with different roles. GM-CSF has a role in Received February 12, 2015, and in revised form, July 24, 2015 inducing emergency hematopoiesis not in steady state, and Published, MCP Papers in Press, July 30, 2015, DOI influences the pathogenesis of various inflammatory as well 10.1074/mcp.M115.048744 as autoimmune diseases (5). In this line, in vitro generated Authorship: Contribution: S.H.S., K.P.K. and Y.R.N. designed the experiments. J.H.H. and J.H.J. performed the proteomic experi- ments. Y.R.N., D.J. and D.S. performed macrophage functional ex- The abbreviations used are: GM-CSF, granulocyte-macrophage periments. M.Y.L. and Y.R.N. analyzed the proteomic data. M.C. colony stimulating factor; M-CSF, macrophage colony stimulating constructed figures and Y.W.K. described the schematic image. factor; DEP, differentially expressed proteins; DPP, differentially Y.R.N. and M.Y.L. wrote the manuscript. phosphorylated proteins; GOBP, Gene Ontology biological process. 2722 Molecular & Cellular Proteomics 14.10 This is an Open Access article under the CC BY license. Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages Animal Care and Use Committee of Seoul National University (acces- GM-CSF grown macrophages are now considered as pro- sion number SNU-130311-2-2). Bone marrow derived macrophages inflammatory macrophages that display a robust immune re- (BMMs) were isolated as described before (16). To enrich the macro- sponses upon LPS stimulation compared with M-CSF grown phage population, we supplemented complete medium with 25 ng/ml macrophages (6). In contrast, M-CSF contributes the mainte- murine GM-CSF (Miltenyi Biotech, Bergisch Gladbach, Germany) for nance of most resident macrophages including osteoclast in GM-CSF grown bone marrow derived macrophages (GM-BMMs) or vivo and is known to affect homeostatic anti-inflammatory 20% L929 murine fibrosarcoma cell line culture supernatants for M-CSF grown bone marrow derived macrophage (M-BMMs). After 7 characteristics of macrophages. M-CSF grown macrophages days of differentiation, GM-BMMs and M-BMMs were stimulated with are widely accepted as in vitro-generated macrophage lipopolysaccharide (E. coli LPS, Sigma) 100 ng/ml for 2 h. Four ex- sources because they showed relatively homogenous and perimental groups (GM-BMM, GM-BMM/LPS, M-BMM, M-BMM/ stable macrophage phenotypes (7). A number of genomic LPS) were processed for further analysis. studies have been performed to analyze macrophage activa- Flow Cytometry—Macrophages were scrapped and incubated for 20 min with appropriate antibodies diluted to optimal concentrations tion in response to pro-inflammatory/anti-inflammatory stim- in FACS buffer (PBS, 5% FBS, 5 mM EDTA, and 1% NaN3). For cell uli. However, to date, there have been no clear reports on the sorting and proteomics analysis, antibodies included anti-CD45 (30- global proteomic differences that govern the functional char- F11), F4/80 (BM8), CD11b (M1/70), and IA/IE (M5/114.15.2). For sur- acteristics of differently differentiated or activated macro- face marker expression analysis, additional antibodies included phages (8–11). To fully elucidate what enables pro-inflamma- CD11c (HL3), CD80 (L307.4), MerTK (clone 125518, R&D systems, tory macrophages to be poised for rapid and robust immune Minneapolis, MN), and CD64 (X54–5/7.1, BD Biosciences, Franklin Lakes, NJ) and all were purchased from eBioscience unless indicated. responses requires assessing their global intracellular pro- Cells were sorted on a FACSAria or data were acquired on a LSRII teomic network signatures. flow cytometer (BD Biosciences). The surface marker expressions There has long been an appreciation, especially in the were analyzed using the FlowJo software. cancer field, for how changes in cellular activation coincide Trypsin Digestion, TMT Labeling, and OFF-Gel Fractionation— with alterations in cellular metabolic states (12, 13). Impor- Equal protein amounts were digested using Filter Aided Sample Prep- aration (FASP) method (17). Briefly, lysates were washed with 8 M tantly, over the last couple of years it is becoming increas- urea, followed by alkylation with 50 mM iodoacetamide (20 min at RT). ingly clear that immune cell activation is also coupled to After alkylation, filters were washed with 8 M Urea two times and with profound changes in cellular metabolism and that their fate 0.1 M TEAB two times for labeling with TMT reagents. Trypsin was and function are metabolically regulated (14). In line with added at a ratio of 1 g trypsin : 50 g protein and samples were this, in this study, we found that GM-CSF grown macro- incubated overnight at 37 °C. Tryptic digests of the four cell lysates phages have a higher glycolytic capacity through up-regu- were labeled with four different mass-tags among the TMT sixplex reagents respectively, per the protocol described by Thermo Fisher lated glycolytic enzymes, as well as high lipid/nitrogen com- Scientific (Rockford, IL). Resulting TMT-labeled peptides were subject pound biosynthetic enzymes compared with M-CSF grown to peptide isoelectrofocusing (IEF) fractionation, the 3100 OFFGEL macrophages. They produce robust inflammatory cytokines fractionator with a “Low Resolution Kit” pH 3–10 (Agilent Technolo- upon TLR ligand stimulation only when sufficient glucose is gies, Santa Clara, CA) according to the manufacturer’s instructions. available. Phosphopeptide Enrichment Using TiO —Phosphopeptide enrich- Tm ment was carried out as described in the Titansphere Phos-TiO Kit Here we performed a quantitative analysis of the pro- 2 manual. Briefly, phosphopeptides were eluted with 5% aqueous am- teome/phosphoproteome of primary GM-CSF and M-CSF monium hydroxide and 5% aqueous pyrrolidine solutions from Phos- grown macrophages using the latest isobaric tag based TiO (3 mg/200 l, Titansphere, GL Sciences Inc, Tokyo, Japan) spin TMT labeling and LC-MS/MS (15). This proteomic approach column. with high throughput technology is the first attempt to show Mass Spectrometric Analysis and Database Search—The extracted tryptic peptides were analyzed using a Q-Exactive mass spectrome- the fundamental differences between primary GM-CSF and ter (Thermo Fisher Scientific, Bremen, Germany) coupled with an M-CSF grown macrophages and finally reveals that innate Easy-nLC system (Thermo Fisher Scientific, Odense, Denmark). Tryp- cellular anabolic metabolism paves the way for inducing tic peptides were resuspended in 0.1% formic acid and separated on robust immune responses. In this study, we describe indi- EASY-Spray column (C18, 2 m particle size, 100 Å pore size, 75 m vidual differentially expressed proteins in the total network id  50 cm length, Thermo Fisher Scientific). Samples were resolved maps of GM-CSF and M-CSF grown macrophages and with a linear gradient of solvent B (100% ACN, 0.1% formic acid); 5–50% over 76 min, 50–90% over 12 min at a flow rate of 300 nL/min. predict how they are specifically involved in initiating inflam- The separated peptide ions eluted from the analytic column were mation or resolution. entered into the mass spectrometer at an electrospray voltage of 2.1 kV. All MS/MS spectra were acquired in a data-dependent mode for EXPERIMENTAL PROCEDURES fragmentation of the ten most abundant peaks from the full MS scan Macrophages Preparation—C57BL/6J mice were obtained from with 30% normalized collision energy. The dynamic exclusion dura- Jackson Laboratory (Bar Harbor, ME). Male mice of five to 10 weeks tion was set at 20 s and the isolation mass width was 2.5 Da. MS of age were used to isolate bone marrow cells. The mice were housed spectra were acquired with a mass range of 400–1800m/z and 70,000 under specific pathogen-free (SPF) conditions and cared according to resolution at m/z 200. MS/MS resolution was acquired at a resolution the Guide for the Care and Use of Laboratory Animals prepared by the of 17,500. The acquired MS/MS spectra were searched against the Institution of Animal Care and Use Committee of Seoul National Universal Protein Resource mouse protein database (Uniprot release University. All of the experiments were approved by the Institute for 2013_09, 50287 entries, http://www.uniprot.org/) with the Sequest Molecular & Cellular Proteomics 14.10 2723 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages algorithm in Proteome Discoverer 1.4 (Thermo Fisher Scientific, Bre- Seahorse cell plates (5  10 cells per well). Before plate reading, men, Germany). Search parameters were as follows: tryptic specificity cells were washed three times with glucose free assay media (Sea- with up to two missed cleavage sites, mass tolerances for precursor horse Bioscience), and the OCR and ECAR were assessed in glu- ions and fragment ions were set to 10 ppm and 0.8 Da, respectively, cose-containing assay media. Perturbation profiling of GM-BMMs fixed modification for carbamidomethyl-cysteine, TMT 4-plex of ly- and M-BMMs metabolic pathways was achieved by the addition of M), oligomycin (5 M), 2-deoxyglucose (100 mM) or LPS. sine and N terminus and variable modification for methionine oxida- glucose (10 m Experiments with the Seahorse system were done with the following tion. Following database searching, the output files were imported assay conditions: 2 min mixture; 2 min wait; and 4–5 min measure- into Scaffold Q (version Scaffold_4.3.2, Proteome Software Inc. ment. Metabolic parameters were then calculated. Portland, OR). Scaffold was used to organize all data, to quantitate Statistical Analysis—All data unless otherwise indicated are shown protein and to validate peptide identifications using the Peptide as mean  S.E. and were tested using two-tailed Student’s t test or Prophet algorithm (18). We selected peptides with cutoff of FDR two-way ANOVA using GraphPad Prism 4. 0.01 and Peptide threshold 95%. We then identified the proteins that have Protein threshold 99.9%. Validated data were normalized be- RESULTS tween intrasample channels and TMT signals showed at least 1.5-fold change in abundance were used for quantitative analysis of protein. Quantitative Analysis of Proteome and Phosphoproteome of The mass spectrometry proteomics data have been deposited to the GM-BMMs and M-BMMs—Macrophage sorting for proteomic ProteomeXchange Consortium (19) via the PRIDE partner repository analysis was performed according to the gating strategy with the dataset identifier PXD002582. shown in Fig. 1A. We first characterized in vitro GM-CSF GO Analyses—Gene Ontology biological processes (GOBPs) rep- resented by the sets of proteins were determined using the DAVID grown bone marrow derived macrophages (GM-BMMs) and software (20). For each set of proteins, we identified the GOBPs M-CSF grown bone marrow derived macrophages (M- represented by the genes with p value  0.05 using STRING v9.1 (21). BMMs). GM-GMMs were CD45 , F4/80 , and MHCII .M- The degree of centrality (K) and shortest-path centrality (SP) were BMMs showed homogenous population and expressed computed using CentiScaPe (22), and the nonseed proteins with CD45, F4/80, and CD11b. GM-BMMs expressed CD11c, SP  0orK  1 were removed. The nodes with the same GOBPs were grouped into the same modules, each of which was named by CD80 as well as the macrophage marker CD64 (Fig. 1B)(26). the corresponding GOBP. The network was visualized using Cyto- In contrast, M-BMMs expressed MHCII, CD11c, and CD80 scape (v. 2.8.3) (23). scarcely, indicating poor antigen presentation ability. Instead, RNA Isolation and Gene Expression Profiling—Global gene expres- they expressed relatively higher MerTK, which was further sion analyses were performed using Affymetrix GeneChip® Mouse up-regulated by M-CSF (27). Further experiments used sorted Gene 1.0 ST oligonucleotide arrays. The sample preparation was GM-BMMs and M-BMMs according to the above gating strat- performed according to the instructions and recommendations pro- vided by the manufacturer. Total RNA was extracted by Trizol reagent egy. In case of GM-CSF cultured cells, small population of (Invitrogen, Carlsbad, CA) according to the manufacturer’s instruc-  low high CD45 F4/80 MHCII dendritic cells were excluded tions. Expression data were generated by Affymetrix Expression Con- throughout the experiments. To enrich early TLR-responsive sole software version1.1. For normalization, RMA (Robust Multi-Av- proteins as well as core polarization specific proteins, we had erage) algorithm was implemented in the Affymetrix Expression four experimental groups of GM-BMMs and M-BMMs, with Console software. Western blotting—Sorted GM-BMMs and M-BMMs were lysed and and without LPS stimulation for 2 h (Fig. 1C). Three biologic analyzed for protein expression as described earlier (24). Antilactate replicates were generated by combining the protein extracts dehydrogenase B chain (Thermo), antiphosphofructokinse-1 (Santa- from five mice per group. The TMT labeling (fourplex) quanti- Cruz, TX), antiribose phosphate isomerase a (Abcam, Cambridge, tative proteomic method was used to profile changes in both UK), antitransferrin receptor (SantaCruz) and antibeta actin (Santa- protein abundance and phosphorylation stoichiometry in the Cruz) were used at 1/2000 dilutions. Cytokine Production—Sorted GM-BMMs and M-BMMs were whole cell extracts. Comparative MS analysis of these ex- seeded in 96-well plates at 5  10 /200 l with complete media and tracts on a Q-Exactive mass spectrometer led to the identifi- rested for 6 h. Macrophages were stimulated with LPS (100 ng/ml) for cation of 33,472 unique peptides from 3990 proteins with an 4 h and 24 h. Supernatants were collected and stored at 80 °C until average sequence coverage of 23% (Fig. 1D, supplemental TNF, IL-6, and IL-10 quantification. For IL-1 quantitation, cells were table S12-S13). Although expression levels remained unaf- lysed in RIPA buffer and stored at 80 °C. Cells in each well were fected by TLR stimulation for most of the identified proteins of analyzed for their protein concentration using the BCA assay kit (Thermo, Waltham, MA) and this was used as a normalization factor GM-BMMs and M-BMMs, a subset of 294 proteins showed for cytokine quantitation. Cytokine concentrations were determined more than a 1.5-fold change in abundance (supplemental using the duoset ELISA kit (R&D systems, Minneapolis, MN). Tables S1 and S2). Latex Bead Phagocytosis—Sorted GM-BMMs and M-BMMs were 6 We next examined the differences in the status of protein seeded in 12-well plates at 10 /ml with complete media and rested for phosphorylation between GM-BMMs and M-BMMs as well as 6 h. Each well of cells were incubated with 10 Alexa 350-tagged latex beads (Molecular Probe, Eugene, OR) of 1 m diameter for 2 h. The GM-BMMs/LPS and M-BMMs/LPS (Fig. 1E). This analysis led number of phagocytozed beads were analyzed on a LSRII flow cy- to the identification of 2239 unique phosphopeptides from tometer (BD Biosciences) and analyzed using the FlowJo software. 1274 phosphoproteins (4844 phosphorylation sites, peptide ECAR and OCR Measurements—The Seahorse XF-24 metabolic probability95% and protein probability99%). A compari- extracellular flux analyzer (Seahorse Bioscience, Billerica, MA) was son of the identified phosphorylation sites with those from used to analyze the ECAR (in mpH/min) and the OCR (in pmol/min) PHOSIDA (28), PhosphoSitePlus (29), and PhosphoELM (30) (25). Briefly, GM-BMMs and M-BMMs were differentiated for 7 days in 2724 Molecular & Cellular Proteomics 14.10 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages FIG.1. Quantitative proteomic/phosphoproteomic analysis of GM-BMMs/M-BMMs. A, GM-BMMs and M-BMMs were differentiated over 7 days from C57BL/6J male mice bone marrow cells with GM-CSF or M-CSF supplementation, respectively. Cells were sorted by CD45 F4/ 80 MHCII population for GM-BMMs, or CD45 F4/80 CD11b population for M-BMMs. B, Surface marker examination of GM-BMMs and M-BMMs. Attached cells were stained with the appropriate antibodies and analyzed for surface MHCII, CD11c, CD80, CD64, and MerTK expressions per gated populations in A. Three independent experiments were performed. **, p  0.01, ***, p  0.001 by Student t test. C, Experimental strategy. Macrophage proteins were pooled to generate three biological replicates for each of the GM-BMM/M-BMM/GM- BMMLPS/M-BMMLPS experimental groups. D, Results of protein expression experiments. E, Results of phosphoproteome experiments. F, Scatter plot of gene symbols detected both in proteome and microarray analyses. Data represents the fold change of log (GM-BMM/M-BMM) of 2491 genes. Regression p value  2.1987e-242, R-squre  0.31352. G, GO analysis of differentially expressed proteins and phosphopeptides. Heat-map shows significant GO biological process terms (p  0.05) for differentially expressed proteins/phosphoproteins between GM-BMM/M- BMM and GM-BMMLPS/M-BMMLPS. Red color indicates increased protein/phosphoprotein abundances in GM-BMM or GM-BMM/LPS, and blue color indicates increased abundances in M-BMM or M-BMM/LPS. Star markings show the most enriched GO category in a heatmap. Molecular & Cellular Proteomics 14.10 2725 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages TABLE I List of proteins enriched in Gene Ontology Biological Process category from differentially expressed protein abundances between GM-BMMs versus M-BMMs and GM-BMMsLPS vs M-BMMsLPS: upregulated in GM-BMMs Gene symbol Protein Name Ratio GM-BMM/M-BMM p value Fold enrichment Glucose metabolic process 0.00006 7.84 ALDOC Fructose-bisphosphate aldolase C 2.22 ENO1 Alpha-enolase 1.57 FABP5 Fatty acid-binding protein, epidermal 2.64 GYS1 Glycogen starch synthase, muscle 1.68 LDHB L-lactate dehydrogenase B chain 2.00 PFKP 6-phosphofructokinase 1.80 PYGL Glycogen phosphorylase, liver form 1.62 RPIA Ribose-5-phosphate isomerase 1.62 Nitrogen compound biosynthetic process 0.02197 3.18 ASS1 Argininosuccinate synthetase 1 3.03 ATP6V0D2 V-type proton ATPase subunit d2 2.93 KYNU Kynureninase 1.62 PADI4 Protein-arginine deiminase type-4 1.74 PHGDH D-3-phosphoglycerate dehydrogenase 2.00 RRM1 Ribonucleoside-diphosphate reductase large subunit 1.93 SLC7A2 Low affinity cationic amino acid transporter 2 1.68 Lipid biosynthetic process 0.00021 4.80 ACLY ATP-citrate synthase 1.62 ALOX5 Arachidonate 5-lipoxygenase 1.68 CD74 H-2 class II histocompatibility antigen gamma chain 2.46 FABP5 Fatty acid-binding protein, epidermal 2.64 FASN Fatty acid synthase 1.62 FDPS Farnesyl pyrophosphate synthase 2.00 HMGCS1 Hydroxymethylglutaryl-CoA synthase, cytoplasmic 1.68 IDI1 Isopentenyl-diphosphate Delta-isomerase 1 1.93 LTA4H Leukotriene A4 hydrolase 2.00 MVD Diphosphomevalonate decarboxylase 2.00 indicated that 2541 sites ( 52%) have not been previously differentially phosphorylated proteins (DPPs) in GM-BMMs/ reported. We also evaluated the confidence level of site lo- M-BMMs and GM-BMMs/LPS M-BMMs/LPS pairs (Fig. 1G). calization using a probability score function and determined A total of 52 categories were enriched and are presented in a that 34% of the identified sites (1635 sites) corresponded to heatmap. Interestingly, the most obvious enriched categories high confidence assignments (supplemental Table S11). in both GM-BMMs and GM-BMMs/LPS macrophages were Among them, 450 and 899 phosphopeptides, corresponding metabolic processes (Table I). These include carbohydrate to 228 and 481 proteins, were differentially regulated between catabolic, glucose metabolic, alcohol catabolic, lipid biosyn- GM-BMMs and M-BMMs upon LPS stimulation, respectively thetic, and amino acid metabolic processes (Table I). The GO (supplemental Table S8). term macromolecular complex subunit organization was en- A scatter plot of protein abundance and mRNA expression hanced in GM-BMMs/LPS compared with GM-BMMs. These (GEO accession number GSE63245) showed a general pre- energy consuming processes are potentially supported by diction ability of microarray data to project protein quantity glucose metabolic/nitrogen compound biosynthetic and lipid (Fig. 1F), but is still limited in representing exact whole protein synthetic pathways in GM-BMMs, as with rapidly proliferating quantities of macrophages with a regression p value of cancer cells (31). Secretion, chromatin organization, and apo- 2.1987e ptosis were specifically up-regulated in GM-BMMs/LPS. and R-square of 0.31352. This means that sub- stantial macrophage proteins go through post-translational In contrast, M-BMMs and M-BMMs/LPS had different en- modification and our proteomics data could give valuable and riched functions, in that they showed higher endocytosis and precise insight into these global differences between opposite homeostasis processes. Of note, “energy derivation by oxi- characterized macrophages. dation of organic compounds” was highly up-regulated in GO Analysis Revealed Metabolism as a Key Regulator of M-BMMs (Fig. 1G), showing a metabolic state that is the Macrophage Function—To gain insight into the functional dif- converse of GM-BMMs (Table II). This decreased upon LPS ferences of each macrophage type, we conducted a compar- stimulation, implying a metabolic shift from oxidation to other ison of Gene Ontology (GO) annotations of biological pro- processes. Subcellular distribution of DEPs also showed that cesses from differentially expressed proteins (DEPs) and mitochondrial components were highly enriched in M-BMMs 2726 Molecular & Cellular Proteomics 14.10 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages TABLE II List of proteins enriched in Gene Ontology Biological Process category from differentially expressed protein abundances between GM-BMMs versus M-BMMs and GM-BMMsLPS vs M-BMMsLPS: upregulated in M-BMMs Fold Gene symbol Protein Name Ratio M-BMM/GM-BMM p value enrichment Energy derivation by oxidation of organic compounds 0.00180 6.82 CAT Catalase 1.80 GAA Lysosomal alpha-glucosidase 1.57 SDHC Succinate dehydrogenase cytochrome b560 subunit, 1.74 mitochondrial SLC37A2 Sugar phosphate exchanger 2 1.74 SOD2 Superoxide dismutase Mn , mitochondrial 2.46 SUCLG2 Succinyl-CoA ligase GDP-forming subunit beta, 1.57 mitochondrial Endocytosis 0.00004 5.9 ABCA1 ATP-binding cassette, sub-family A (ABC1), member 1 1.57 ARHGAP27 Rho GTPase activating protein 27 1.57 BET1 Bet1 golgi vesicular membrane trafficking protein 1.52 CD36 CD36 molecule (thrombospondin receptor) 1.80 EHD1 EH-domain containing 1 2.22 ELMO1 engulfment and cell motility 1 1.52 FNBP1L formin binding protein 1-like 2.00 HCK hemopoietic cell kinase 1.74 ITSN1 intersectin 1 (SH3 domain protein) 2.14 TFRC transferrin receptor 2.46 VAV1 vav 1 guanine nucleotide exchange factor 1.57 compared to GM-BMMs, suggesting a bias for oxidation pro- that is indirectly related to glycolysis (Fig. 2A). Among these, cesses as their cellular energy source (supplemental Table we confirmed the abundant protein expressions of ribose 5 S9). Several signaling pathways related to endocytosis/phag- phosphate isomerase (RPIA) and 6-phosphofructokinase ocytosis, including small GTPase and Ras, were enriched in (PFKP) in GM-BMMs compared with M-BMMs by Western both M-BMMs and M-BMMs/LPS. blot, indicating the reliability of our proteomic methodology Similar results were also obtained in our integrated pathway (Fig. 2B). analysis (IPA) of DPPs between GM-BMMs and GM-BMMs/ Next we hypothesized that GM-BMMs have a much higher LPS or M-BMMs and M-BMMs/LPS. GM-BMMs/LPS showed maximum glycolytic capacity than M-BMMs. To assess met- enhanced protein phosphorylations for cytokine signaling, abolic differences, we recorded extracellular acidification rate MAPK signaling, TLR4 cascade as well as mTOR signaling (ECAR) and oxygen consumption rate (OCR) of GM-BMMs events compared with basal GM-BMMs (supplemental Table and M-BMMs. GM-BMMs had significantly higher basal S4). In line with this, 58 proteins were up-regulated in GM- ECAR than M-BMMs, although OCR was similar (Fig. 2C). BMMs/LPS compared with GM-BMMs, mostly related to im- Next we treated the macrophages with glucose and oligomy- mune responses including IL-1, lactotransferrin, TNF, cop- cin to maximize glycolysis as shown in Fig. 2D. This con- per transporter protein ATOX-1, and IL-1 (supplemental firmed that GM-BMMs have a higher glycolytic capacity Table S3). M-BMMs/LPS showed enhanced Rho GTPase and than M-BMMs (Fig. 2E). This difference contributes to the RAC1 activity in IPA of DPPs, which correlate with endocytic varied extent of acute glycolysis upon LPS stimulation, with activity. M-BMMs/LPS showed only seven up-regulated pro- GM-BMMs showing a more prominent and prolonged ECAR teins relative to basal M-BMMs (supplemental Table S7), sug- curve than M-BMMs after LPS treatment (Fig. 2F). Acute gesting remarkable down-regulated immune responses com- glycolysis in macrophages following LPS was totally de- pared with GM-BMMs. Taken together, GM-BMMs had pendent on glucose uptake as ECAR does not increase enhanced anabolic metabolism and responded to LPS with without glucose in the media (Fig. 2G). Taken together, inflammatory activities. On the contrary, M-BMMs had en- these results indicate that GM-CSF grown macrophages hanced endocytic processes and did not show marked in- have inherently higher glycolytic capacities than M-CSF flammatory activities upon LPS stimulation. grown macrophages to enable heightened glucose conver- Glycolysis Signatures GM-CSF Grown Macrophages— sion to lactate in response to LPS. Among the eight proteins involved in glucose metabolism Metabolic Differences are Linked to Cytokine Responses in up-regulated in GM-BMMs, five were direct glycolytic en- Macrophages—Finally, to test the impact of metabolic con- zymes (RPIA, PFKP, LDHB, ENO1, ALDOC) and one, PYGL, struct of GM-BMMs and M-BMMs on cytokine production, Molecular & Cellular Proteomics 14.10 2727 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages FIG.2. GM-BMMs have higher glycolytic capacities than M-BMMs. A, Expression profiles of glycolytic genes and proteins in GM-BMMs compared with M-BMMs. ***, p  0.001 by two-way ANOVA. B, Western blot results for RPIA, PFKP, and actin. C, Basal OCR and ECAR of GM-BMMs and M-BMMs. Data obtained using XF-24 extracellular flux analyzer. Three independent experiments were performed. **, p  0.01 by Student t test. D, Maximum glycolytic capacities of GM-BMMs and M-BMMs. Perturbation profiling of lactate production (ECAR) was achieved by the addition of glucose (10 mM), oligomycin (5 M), and 2-deoxyglucose (2-DG, 100 mM). Three independent experiments were performed. ***, p  0.001 by two-way ANOVA. E, Quantitative graph representing mean maximum glycolytic capacities of GM-BMMs and M-BMMs. Three independent experiments were performed. *, p  0.05 by Student t test. F, LPS induced acute glycolysis. GM-BMMs and M-BMMs were stimulated with 100 ng/ml E. coli LPS and recorded for the ECAR changes. Data showing relative ECAR (%) changes from baseline. Three independent experiments were performed. ***, p  0.001 by two-way ANOVA. G, Glucose effect on acute glycolysis upon LPS stimulation. GM-BMMs were stimulated with LPS with or without media glucose and ECAR changes were recorded. H–K, GM-BMMs produced more inflammatory cytokines than M-BMMs in a glycolysis-dependent manner. GM-BMMs and M-BMMs were sorted and replated in 96-well plates at 5  10 density. Cells were treated with LPS (100 ng/ml) with or without 2-DG (100 mM) and cytokines were measured at 4 h (TNF) and 24 h (IL-6, IL-1, IL-10). ELISA was performed to quantify TNF (H), IL-6 (I), IL-10 (K) production in culture supernatants, or IL-1 synthesis (J) in cell lysates. Three independent experiments were performed. **, p  0.01, ***, p  0.001 by two-way ANOVA. we employed 2-deoxyglucose (2-DG) to interfere with glu- upregulated endocytosis-relating proteins in M-BMMs cose metabolism and measured cytokine production as a (supplemental Table S5). Among the upregulated 11 proteins functional readout. Consistent with previous reports (6), in M-BMMs compared to GM-BMMs, we confirmed that GM-BMMs produced more TNF, IL-6, and IL-1 than M- transferrin receptor, CD71 (TFRC), which transports trans- BMMs (Fig. 2H,2I, and 2J). Importantly, blockade of glycol- ferrin inside the cell and is up-regulated by CSF1 (32), ysis significantly reduced LPS-induced TNF, IL-6, and expression was higher in M-BMMs than GM-BMMs (Fig. IL-1 production to that of basal levels both in GM-BMMs 3B). M-BMMs showed enhanced endocytic processes both and M-BMMs. Both macrophages produced similar levels of in protein expression as well as phosphoprotein enrichment. IL-10 upon LPS stimulation, but this was also inhibited by Thus, we performed latex bead uptake experiments (Fig. 2-DG (Fig. 2K). These results confirm the vital role of gly- 3C). After GM-BMMs and M-BMMs sorting according to colysis in LPS induced pro-inflammatory cytokine re- Fig. 1A, replated macrophages were incubated with Ax350- sponses. Given that LPS induced cytokine production is tagged latex beads of 1 m diameter for 2 h. As expected largely dependent on glycolysis, it goes to reason that the 40% of M-BMMs had more than four beads per cell, but more glycolytically active GM-BMMs can synthesize con- only 20% of GM-BMMs did (Fig. 3D). These results con- siderably more inflammatory cytokines than homeostatic firmed the relationship between our proteome/phosphopro- M-BMMs. teome analysis and macrophage functional characteristics Endocytosis Predominates in M-CSF Grown Macrophag- described above. es—Eleven proteins related to endocytosis were up-regulated Protein Interaction Network Analysis of Macrophages—To in M-BMMs compared to GM-BMMs (Fig. 3A). LPS further gain insight into the proteome and phosphoproteome of GM- 2728 Molecular & Cellular Proteomics 14.10 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages FIG.3. M-BMMs phagocytosed more beads than GM-BMMs. A, Relative mRNA and protein expression levels of endocytosis participants in M-BMMs compared with GM-BMMs. **, p  0.01, ***, p  0.001 by two-way ANOVA. B, Western blot results for TFRC and actin. C, Histograms showing the number of beads per cell. Sorted macrophages were replated and incubated with Alexa350 tagged latex beads for 2 h. Bead phagocytosis was analyzed per macrophage population (GM-BMMs: F4/80 MHCII , M-BMMs: F4/80 CD11b ) using FACS. Cell percentages containing more than four beads are indicated in both histograms. D, Quantitative graph showing cell percentages with more than four beads per cells. Three independent experiments were performed. *, p  0.05 by Student t test. BMMs and M-BMMs, we conducted an Ingenuity Pathway replication. In M-CSF grown macrophages, Rho/GTPase sig- Analysis (IPA) using DEPs and DPPs involved in differentially naling is well interconnected with endocytosis and adhesion. regulated biological processes of six major groups between DISCUSSION GM-BMMs and M-BMMs: (1) Glucose metabolism, (2) Lipid metabolism, (3) Amino acid metabolism, (4) DNA replication, In this study, we used the latest isobaric tag based mul- (5) Endocytosis, and (6) Rho/GTPase signaling. The total net- tiplex proteomic/phosphoproteomic quantitative analysis work map is presented in Fig. 4, which depicts 145 DEPs or method to identify key molecular differences between GM- DPPs in colored figures as well as predicted regulating tran- CSF and M-CSF grown macrophages. We found that GM- scription factors, transporters, enzymes, and signaling mole- BMMs have potentiative glycolytic/lipid biosynthetic path- cules denoted with white circles. Of note, transcription factors ways as well as nitrogen compound biosynthesis processes. that regulate glycolysis such as HIF1A, MYC, FOXO1, FOXA2, These enhanced anabolic pathways directly link to their pro- ATF4, HNF1A, and HNF1B, underpin various other processes inflammatory cytokine production capacity. We also pre- in GM-CSF grown macrophages including glucose/lipid/ sented valuable targets for disrupting macrophage functions, amino acid metabolism, immune response as well as DNA as well as novel phosphosites of macrophage proteins Molecular & Cellular Proteomics 14.10 2729 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages FIG.4. Regulated protein networks in GM-BMMs versus M-BMMs. Protein abundance and phosphorylation data were analyzed with the Ingenuity Pathway Analysis (IPA) using 145 differentially expressed proteins and phosphoproteins involved in differentially regulated biological process between GM-BMMs and M-BMMs. Glucose metabolism, lipid metabolism, amino acid metabolism, DNA replication, endocytosis, and Rho/GTPase signaling were the subnetworks most significantly affected. Proteins/nodes are grouped according to their function. Red color indicates GM-BMMs up-regulation and green is M-BMMs up-regulation. Circle means DEP, diamond means DPP. White circles are predicted interacting signaling molecules as well as transcription factors. All indicated by gene symbol. (supplemental Table S11). To our knowledge, this is the first Of note, the role of glycolysis in GM-BMMs differs from the large scale proteome/phosphoproteome analysis of GM-CSF acute glycolytic switch following LPS stimulation in dendritic and M-CSF grown macrophages. Our reliable data gives in- cells, which results from IKK/TBK1 mediated Akt activation sight into the importance of cellular metabolic polarization and (37), because basal enhanced glycolytic state exists in GM- their functional outcomes in macrophages. CSF primed macrophages in the absence of TLR signaling. It Aberrant carbohydrate metabolism is classically a distinc- also appears that this bias toward glycolysis does not stem tive feature of tumor cells (33). In general, normal cells pro- from TLR mediated inducible nitric oxide synthase (iNOS) duce most of their ATP from glucose through oxidative phos- expression, which produces nitric oxide (NO) and inhibit oxi- phorylation. However, many cancer cells produce ATP by dative phosphorylation (38). Instead, our network analysis conversion of glucose to lactate and therefore exhibit lower predicts that GM-BMMs have up-regulated upstream glucose oxidative phosphorylation activity. This “glycolytic pheno- regulatory factors such as HIF1A, MYC, HNF1A (39), or type” ensures sufficient macromolecule biosynthesis neces- FOXA2 (40). Given that GM-CSF is known to increase L-Myc sary for rapid cell growth and division (34). Compellingly, we expression in dendritic cells after 24 h (41), glycolysis in found that GM-BMMs showed a metabolic state similar to GM-CSF grown macrophages may be possibly further up- cancer cells. We confirmed that the up-regulated glycolytic regulated by Myc because many glycolytic enzymes have enzymes are collectively involved in LPS-induced glycolysis myc-binding sites in their promoters (42). In 1995, Brissette et and consequently connected with inflammatory cytokine pro- al. reported that GM-CSF primes mice for enhanced cytokine ductions. These data give us profound implications in that production in response to LPS, and our results suggest a GM-CSF affected macrophages are usually generated when possible underlying mechanism (43). our body has been influenced by microbial or aseptic inflam- Another interesting result was that GM-BMMs have up- matory stimuli (35). Given that GM-CSF therapy has been regulated enzymes in the mevalonate pathway. They are in- considered to alleviate inflammatory conditions such as ar- cluded in the GO term lipid biosynthetic process: diphospho- thritis (36), enhanced glycolytic pathway of GM-CSF grown mevalonate decarboxylase (MVD), farnesyl pyrophosphate macrophages might be an alternative specific valuable target synthase (FDPS), hydroxymethylglutaryl-CoA synthase for macrophage mediated inflammatory diseases. (HMGCS1), and isopentenyl-diphosphate delta-isomerase 1 2730 Molecular & Cellular Proteomics 14.10 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages (IDI1) (Table I). Similar to glycolysis, the mevalonate pathway granulocyte-macrophage colony-stimulating factor. Crit. Rev. Immunol. 25, 405–428 is dysregulated in tumor cells (44) and several researchers 5. Hamilton, J. A. (2008) Colony-stimulating factors in inflammation and au- have tried to target this pathway to inhibit cancer cell malig- toimmunity. Nat. Rev. Immunol. 8, 533–544 nancy (45). The mevalonate pathway is a complex biochem- 6. Fleetwood, A. J., Lawrence, T., Hamilton, J. A., and Cook, A. D. (2007) Granulocyte-macrophage colony-stimulating factor (CSF) and macro- ical pathway that generates several fundamental end-prod- phage CSF-dependent macrophage phenotypes display differences in ucts including cholesterol, isoprenoids, dolichol, ubiquinone, cytokine profiles and transcription factor activities: implications for CSF and isopentenyladenine (45). In fact, glycolysis cascades into blockade in inflammation. J. Immunol. 178, 5245–5252 7. Murray, P. J., Allen, J. E., Biswas, S. K., Fisher, E. A., Gilroy, D. W., Goerdt, the mevalonate pathway when macrophages are activated S., Gordon, S., Hamilton, J. A., Ivashkiv, L. B., Lawrence, T., Locati, M., because glycolysis induced by TLR activation produces Mantovani, A., Martinez, F. O., Mege, J. L., Mosser, D. M., Natoli, G., abundant pyruvate and acetyl-CoA followed by lipid accumu- Saeij, J. P., Schultze, J. L., Shirey, K. A., Sica, A., Suttles, J., Udalova, I., van Ginderachter, J. A., Vogel, S. N., and Wynn, T. A. (2014) Macrophage lation inside cells (46). Thus, it would be interesting to inves- activation and polarization: nomenclature and experimental guidelines. tigate whether the mevalonate pathway contributes to the Immunity 41, 14–20 macrophage pro-inflammatory phenotype. 8. Martinez, F. O., Gordon, S., Locati, M., and Mantovani, A. (2006) Transcrip- tional profiling of the human monocyte-to-macrophage differentiation The immune system is comprised of a heterogeneous pop- and polarization: new molecules and patterns of gene expression. J. Im- ulation of cells that, for the most part, are relatively quiescent munol. 177, 7303–7311 in the steady state but share the ability to rapidly respond to 9. McDermott, J. E., Archuleta, M., Thrall, B. D., Adkins, J. N., and Waters, K. M. (2011) Controlling the response: predictive modeling of a highly infection, inflammation, and other perturbations. There is a central, pathogen-targeted core response module in macrophage acti- growing appreciation of the fact that transitions between qui- vation. PloS One 6, e14673 escent and activated states require the apportioning of nutri- 10. Nau, G. J., Richmond, J. F., Schlesinger, A., Jennings, E. G., Lander, E. S., and Young, R. A. (2002) Human macrophage activation programs in- ents into different pathways, and, therefore, there is a strong duced by bacterial pathogens. Proc. Natl. Acad. Sci. U.S.A. 99, interest in how metabolic pathways are regulated to support 1503–1508 or direct functional changes. Several reports have been al- 11. Ramsey, S. A., Klemm, S. L., Zak, D. E., Kennedy, K. A., Thorsson, V., Li, ready noted the importance of macrophage metabolic states B., Gilchrist, M., Gold, E. S., Johnson, C. D., Litvak, V., Navarro, G., Roach, J. C., Rosenberger, C. M., Rust, A. G., Yudkovsky, N., Aderem, on polarization and attempted to target various metabolic A., and Shmulevich, I. (2008) Uncovering a macrophage transcriptional pathways, such as glucose-6-phosphate dehydrogenase (47), program by integrating evidence from motif scanning and expression CARKL (48), PGC-1 dynamics. PLoS Comput. Biol. 4, e1000021 (49), TRAP1 (50), or glucose transporter 12. Vander Heiden, M. G., Cantley, L. C., and Thompson, C. B. (2009) Under- 1(51), to modulate macrophage phenotype. Our data clearly standing the Warburg effect: the metabolic requirements of cell prolifer- reveals a systemic link between macrophage activation and ation. Science 324, 1029–1033 metabolic state. In addition, this study identifies novel oppor- 13. Raffaello, A. and Rizzuto, R. (2011) Mitochondrial longevity pathways. Biochim Biophys Acta. 1813, 260–268 tunities for targeting macrophage mediated immune re- 14. Pearce, E. L., and Pearce, E. J. (2013) Metabolic pathways in immune cell sponses. activation and quiescence. Immunity 38, 633–643 15. Dayon, L., and Sanchez, J. C. (2012) Relative protein quantification by * This work was supported by Basic Science Research Program MS/MS using the tandem mass tag technology. Methods Mol. Biol. 893, through the National Research Foundation of Korea (NRF) funded 115–127 by the Ministry of Science, ICT & Future Planning (NRF- 16. Na, Y. R., Yoon, Y. N., Son, D. I., and Seok, S. H. (2013) Cyclooxygenase-2 2014R1A1A1008012, NRF-2012R1A1A3013393) and supported by inhibition blocks M2 macrophage differentiation and suppresses metas- tasis in murine breast cancer model. PloS One 8, e63451 Proteogenomic Research Program, and the Bio- and Medical Tech- 17. AM, V. O. s., Dominin, S. G., Kutepov, E. N., Leonov, A. V., Lobov, A. V., nology Development Program (Project No. 2012M3A9B6055305) Maimulov, V. G., Nesvizhskii Iu, V., Semenova, V. V., Fokin, M. V., and through the National Research Foundation of Korea funded by the Tselykovskaia, N. (2003) [Training physicians in medical prevention spe- Korean Ministry of Education, Science and Technology (to K.P.K.). cialty: problems and prospects]. Gigiena i Sanitariia, 1, 13–15 □ S This article contains supplemental Tables S1 to S13. 18. Keller, A., Nesvizhskii, A. I., Kolker, E., and Aebersold, R. (2002) Empirical ** These authors contributed equally to this work. statistical model to estimate the accuracy of peptide identifications made To whom correspondence should be addressed: Kyung Hee by MS/MS and database search. Anal. Chem. 74, 5383–5392 University, Dept. of Applied Chemistry, Rm471-4 Engineering 19. Vizcaino, J. A., Deutsch, E. W., Wang, R., Csordas, A., Reisinger, F., Ríos, BLDG 1732 Deokyeong-daero, Giheung-gu, Yongin 446-701, Re- D., Dianes, J. A., Sun, Z., Farrah, T., Bandeira, N., Binz, P. A., Xenarios, public of Korea. Tel.: 82-31-201-3868; Fax: 82-31-201-2825; E- I., Eisenacher, M., Mayer, G., Gatto, L., Campos, A., Chalkley, R. J., Kraus, H. J., Albar, J. P., Martinez-Bartolome, S., Apweiler, R., Omenn, mail: [email protected]. Seoul National University College of Medi- G. S., Martens, L., Jones, A. R., and Hermjakob, H. (2014) ProteomeX- cine, Dept. of Microbiology and Immunology, and Institute of change provides globally coordinated proteomics data submission and endemic disease, Seoul 110-799, Republic of Korea. Tel.: 82-2- dissemination. Nat. Biotechnol. 32, 223–226 740-8302; Fax: 82-2-743-0881; E-mail: [email protected]. 20. Huang da, W., Sherman, B. T., and Lempicki, R. A. (2009) Systematic and integrative analysis of large gene lists using DAVID bioinformatics re- REFERENCES sources. Nat. Protoc. 4, 44–57 1. Gordon, S., and Taylor, P. R. (2005) Monocyte and macrophage heteroge- 21. Franceschini, A., Szklarczyk, D., Frankild, S., Kuhn, M., Simonovic, M., Roth, A., Lin, J., Minguez, P., Bork, P., von Mering, C., and Jensen, L. J. neity. Nat. Rev. Immunol. 5, 953–964 (2013) STRING v9.1: protein–protein interaction networks, with increased 2. Mosser, D. M., and Edwards, J. P. (2008) Exploring the full spectrum of coverage and integration. Nucleic Acids Res. 41, D808–815 macrophage activation. Nat. Rev. Immunol. 8, 958–969 22. Scardoni, G., Petterlini, M., and Laudanna, C. (2009) Analyzing biological 3. Gordon, S. (2003) Alternative activation of macrophages. Nat. Rev. Immu- network parameters with CentiScaPe. Bioinformatics 25, 2857–2859 nol. 3, 23–35 4. Fleetwood, A. J., Cook, A. D., and Hamilton, J. A. (2005) Functions of 23. Smoot, M. E., Ono, K., Ruscheinski, J., Wang, P. L., and Ideker, T. (2011) Molecular & Cellular Proteomics 14.10 2731 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages Cytoscape 2.8: new features for data integration and network visualiza- 39. Wang, H., Antinozzi, P. A., Hagenfeldt, K. A., Maechler, P., and Wollheim, tion. Bioinformatics 27, 431–432 C. B. (2000) Molecular targets of a human HNF1 alpha mutation respon- 24. Son, D., Na, Y. R., Hwang, E. S., and Seok, S. H. (2014) Platelet-derived sible for pancreatic beta-cell dysfunction. EMBO J. 19, 4257–4264 growth factor-C (PDGF-C) induces anti-apoptotic effects on macro- 40. Wolfrum, C., Shih, D. Q., Kuwajima, S., Norris, A. W., Kahn, C. R., and phages through Akt and Bad phosphorylation. J. Biol. Chem. 289, Stoffel, M. (2003) Role of Foxa-2 in adipocyte metabolism and differen- 6225–6235 tiation. J. Clin. Invest. 112, 345–356 25. Spolarics, Z., and Wu, J. X. (1997) Role of glutathione and catalase in H2O2 41. Kc, W., Satpathy, A. T., Rapaport, A. S., Brisen˜ o, C. G., Wu, X., Albring, detoxification in LPS-activated hepatic endothelial and Kupffer cells. J. C., Russler-Germain, E. V., Kretzer, N. M., Durai, V., Persaud, S. P., Am. J. Physiol. 273, G1304–1311 Edelson, B. T., Loschko, J., Cella, M., Allen, P. M., Nussenzweig, M. C., 26. Gautier, E. L., Shay, T., Miller, J., Greter, M., Jakubzick, C., Ivanov, S., Helft, Colonna, M., Sleckman, B. P., Murphy, T. L., and Murphy, K. M. (2014) J., Chow, A., Elpek, K. G., Gordonov, S., Mazloom, A. R., Ma’ayan, A., L-Myc expression by dendritic cells is required for optimal T-cell priming. Chua, W. J., Hansen, T. H., Turley, S. J., Merad, M., Randolph, G. J., and Nature 507, 243–247 Immunological Genome, C. (2012) Gene-expression profiles and tran- 42. Gordan, J. D., Thompson, C. B., and Simon, M. C. (2007) HIF and c-Myc: scriptional regulatory pathways that underlie the identity and diversity of sibling rivals for control of cancer cell metabolism and proliferation. mouse tissue macrophages. Nat. Immunol. 13, 1118–1128 Cancer Cell 12, 108–113 27. Zizzo, G., Hilliard, B. A., Monestier, M., and Cohen, P. L. (2012) Efficient 43. Brissette, W. H., Baker, D. A., Stam, E. J., Umland, J. P., and Griffiths, R. J. clearance of early apoptotic cells by human macrophages requires M2c (1995) GM-CSF rapidly primes mice for enhanced cytokine production in polarization and MerTK induction. J. Immunol. 189, 3508–3520 response to LPS and TNF. Cytokine 7, 291–295 28. Gnad, F., Gunawardena, J., and Mann, M. (2011) PHOSIDA 2011: the 44. Clendening, J. W., Pandyra, A., Boutros, P. C., El Ghamrasni, S., Khosravi, post-translational modification database. Nucleic Acids Res. 39, F., Trentin, G. A., Martirosyan, A., Hakem, A., Hakem, R., Jurisica, I., and D253–260 Penn, L. Z. (2010) Dysregulation of the mevalonate pathway promotes 29. Hornbeck, P. V., Kornhauser, J. M., Tkachev, S., Zhang, B., Skrzypek, E., transformation. Proc. Natl. Acad. Sci. U.S.A. 107, 15051–15056 Murray, B., Latham, V., and Sullivan, M. (2012) PhosphoSitePlus: a 45. Swanson, K. M., and Hohl, R. J. (2006) Anti-cancer therapy: targeting the comprehensive resource for investigating the structure and function of mevalonate pathway. Curr. Cancer Drug Targets 6, 15–37 experimentally determined post-translational modifications in man and 46. Feingold, K. R., Shigenaga, J. K., Kazemi, M. R., McDonald, C. M., Patzek, mouse. Nucleic Acids Res. 40, D261–270 S. M., Cross, A. S., Moser, A., and Grunfeld, C. (2012) Mechanisms of 30. Dinkel, H., Chica, C., Via, A., Gould, C. M., Jensen, L. J., Gibson, T. J., and triglyceride accumulation in activated macrophages. J. Leukocyte Biol. Diella, F. (2011) Phospho.ELM: a database of phosphorylation 92, 829–839 sites–update 2011. Nucleic Acids Res. 39, D261–267 47. Ham, M., Lee, J. W., Choi, A. H., Jang, H., Choi, G., Park, J., Kozuka, C., 31. Imamura, K., Ogura, T., Kishimoto, A., Kaminishi, M., and Esumi, H. (2001) Sears, D. D., Masuzaki, H., and Kim, J. B. (2013) Macrophage glucose- Cell cycle regulation via p53 phosphorylation by a 5 -AMP activated 6-phosphate dehydrogenase stimulates proinflammatory responses with protein kinase activator, 5-aminoimidazole- 4-carboxamide-1-beta-D- oxidative stress. Mol. Cell. Biol. 33, 2425–2435 ribofuranoside, in a human hepatocellular carcinoma cell line. Biochem. 48. Haschemi, A., Kosma, P., Gille, L., Evans, C. R., Burant, C. F., Starkl, P., Biophys. Res. Commun. 287, 562–567 Knapp, B., Haas, R., Schmid, J. A., Jandl, C., Amir, S., Lubec, G., Park, 32. Lokeshwar, B. L., and Lin, H. S. (1990) Growth factor-dependent regulation J., Esterbauer, H., Bilban, M., Brizuela, L., Pospisilik, J. A., Otterbein, of transferrin receptor in proliferating and quiescent macrophages. Cell. L. E., and Wagner, O. (2012) The sedoheptulose kinase CARKL directs Immunol. 130, 401–415 macrophage polarization through control of glucose metabolism. Cell 33. Rolfe, D. F., and Brown, G. C. (1997) Cellular energy utilization and molec- Metab. 15, 813–826 ular origin of standard metabolic rate in mammals. Physiol. Rev. 77, 49. Vats, D., Mukundan, L., Odegaard, J. I., Zhang, L., Smith, K. L., Morel, 731–758 C. R., Wagner, R. A., Greaves, D. R., Murray, P. J., and Chawla, A. (2006) 34. Mazurek, S., and Eigenbrodt, E. (2003) The tumor metabolome. Anticancer Oxidative metabolism and PGC-1beta attenuate macrophage-mediated Res. 23, 1149–1154 inflammation. Cell Metab. 4, 13–24 35. Becker, L., Liu, N. C., Averill, M. M., Yuan, W., Pamir, N., Peng, Y., Irwin, 50. Yoshida, S., Tsutsumi, S., Muhlebach, G., Sourbier, C., Lee, M. J., Lee, S., A. D., Fu, X., Bornfeldt, K. E., and Heinecke, J. W. (2012) Unique pro- Vartholomaiou, E., Tatokoro, M., Beebe, K., Miyajima, N., Mohney, R. P., teomic signatures distinguish macrophages and dendritic cells. PloS Chen, Y., Hasumi, H., Xu, W., Fukushima, H., Nakamura, K., Koga, F., One 7, e33297 Kihara, K., Trepel, J., Picard, D., and Neckers, L. (2013) Molecular 36. Cook, A. D., Pobjoy, J., Steidl, S., Du¨ rr, M., Braine, E. L., Turner, A. L., chaperone TRAP1 regulates a metabolic switch between mitochondrial Lacey, D. C., and Hamilton, J. A. (2012) Granulocyte-macrophage colo- respiration and aerobic glycolysis. Proc. Natl. Acad. Sci. U.S.A. 110, ny-stimulating factor is a key mediator in experimental osteoarthritis pain E1604–1612 and disease development. Arthritis Res. Ther. 14, R199 51. Freemerman, A. J., Johnson, A. R., Sacks, G. N., Milner, J. J., Kirk, E. L., 37. O’Neill, L. A. (2014) Glycolytic reprogramming by TLRs in dendritic cells. Troester, M. A., Macintyre, A. N., Goraksha-Hicks, P., Rathmell, J. C., Nat. Immunol. 15, 314–315 and Makowski, L. (2014) Metabolic reprogramming of macrophages: 38. Everts, B., and Pearce, E. J. (2014) Metabolic control of dendritic cell glucose transporter 1 (GLUT1)-mediated glucose metabolism drives a activation and function: recent advances and clinical implications. Front. Immunol. 5, 203 proinflammatory phenotype. J. Biol. Chem. 289, 7884–7896 2732 Molecular & Cellular Proteomics 14.10 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Molecular & Cellular Proteomics American Society for Biochemistry and Molecular Biology

Proteomic Analysis Reveals Distinct Metabolic Differences Between Granulocyte-Macrophage Colony Stimulating Factor (GM-CSF) and Macrophage Colony Stimulating Factor (M-CSF) Grown Macrophages Derived from Murine Bone Marrow Cells*

Download PDF
11 pages

Loading next page...
 
/lp/american-society-for-biochemistry-and-molecular-biology/proteomic-analysis-reveals-distinct-metabolic-differences-between-YFlWKWY1OU

References (58)

Publisher
American Society for Biochemistry and Molecular Biology
Copyright
Copyright © 2015 Elsevier Inc.
ISSN
1535-9476
eISSN
1535-9484
DOI
10.1074/mcp.m115.048744
Publisher site
See Article on Publisher Site

Abstract

crossmark Research © 2015 by The American Society for Biochemistry and Molecular Biology, Inc. This paper is available on line at http://www.mcponline.org Proteomic Analysis Reveals Distinct Metabolic Differences Between Granulocyte-Macrophage Colony Stimulating Factor (GM-CSF) and Macrophage Colony Stimulating Factor (M-CSF) Grown Macrophages Derived from □ S Murine Bone Marrow Cells* Yi Rang Na‡**, Ji Hye Hong§**, Min Yong Lee§, Jae Hun Jung§, Daun Jung‡, Kwang Pyo Kim§, Young Won Kim‡, Dain Son‡, Murim Choi¶, and Seung Hyeok Seok II‡ Macrophages are crucial in controlling infectious agents ulation in GM-BMMs depends on their acute glycolytic and tissue homeostasis. Macrophages require a wide capacity. In contrast, M-BMMs up-regulate proteins in- range of functional capabilities in order to fulfill distinct volved in endocytosis, which correlates with a tendency roles in our body, one being rapid and robust immune toward homeostatic functions such as scavenging cellu- responses. To gain insight into macrophage plasticity and lar debris. Together, our data describes a proteomic the key regulatory protein networks governing their spe- network that underlies the pro-inflammatory actions of cific functions, we performed quantitative analyses of the GM-BMMs as well as the homeostatic functions of proteome and phosphoproteome of murine primary GM- M-BMMs. Molecular & Cellular Proteomics 14: 10.1074/ CSF and M-CSF grown bone marrow derived macro- mcp.M115.048744, 2722–2732, 2015. phages (GM-BMMs and M-BMMs, respectively) using the latest isobaric tag based tandem mass tag (TMT) labeling and liquid chromatography-tandem mass spectrometry Macrophages are a heterogeneous population of immune (LC-MS/MS). Strikingly, metabolic processes emerged as cells that are essential for the initiation and resolution of a major difference between these macrophages. Specifi- pathogen- or tissue damage-induced inflammation (1). They cally, GM-BMMs show significant enrichment of proteins show remarkable plasticity that allows them to respond effi- involving glycolysis, the mevalonate pathway, and nitro- ciently to environmental signals and change their phenotype gen compound biosynthesis. This evidence of enhanced and physiology upon cytokine and microbial signaling (2). glycolytic capability in GM-BMMs is particularly signifi- These changes can give rise to populations of cells with cant regarding their pro-inflammatory responses, be- distinct functions that are phenotypically characterized by the cause increased production of cytokines upon LPS stim- production of pro-inflammatory and anti-inflammatory cyto- kines (3). Among the growth factors that affect macrophage activation states, two cytokines that appear to be important in From the ‡Department of Microbiology and Immunology, and In- stitute of Endemic Disease, Seoul National University College of controlling the functions of macrophage lineage populations Medicine, 103 Daehak-ro, Chongno-gu, Seoul 110-799, South Korea; in inflammatory conditions are granulocyte-macrophage col- §Department of Applied Chemistry, Kyung Hee University, 1732 Deo- ony stimulating factor (GM-CSF) and macrophage colony gyeong-daero, Giheung-gu, Yongin-si, Gyeonggi-do 446-701, South stimulating factor (M-CSF) (4). These CSFs are critical to the Korea; ¶Department of Biomedical Science, Seoul National University proper maintenance of steady-state macrophage develop- College of Medicine, 103 Daehak-ro, Chongno-gu, Seoul 110-799, South Korea ment, although with different roles. GM-CSF has a role in Received February 12, 2015, and in revised form, July 24, 2015 inducing emergency hematopoiesis not in steady state, and Published, MCP Papers in Press, July 30, 2015, DOI influences the pathogenesis of various inflammatory as well 10.1074/mcp.M115.048744 as autoimmune diseases (5). In this line, in vitro generated Authorship: Contribution: S.H.S., K.P.K. and Y.R.N. designed the experiments. J.H.H. and J.H.J. performed the proteomic experi- ments. Y.R.N., D.J. and D.S. performed macrophage functional ex- The abbreviations used are: GM-CSF, granulocyte-macrophage periments. M.Y.L. and Y.R.N. analyzed the proteomic data. M.C. colony stimulating factor; M-CSF, macrophage colony stimulating constructed figures and Y.W.K. described the schematic image. factor; DEP, differentially expressed proteins; DPP, differentially Y.R.N. and M.Y.L. wrote the manuscript. phosphorylated proteins; GOBP, Gene Ontology biological process. 2722 Molecular & Cellular Proteomics 14.10 This is an Open Access article under the CC BY license. Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages Animal Care and Use Committee of Seoul National University (acces- GM-CSF grown macrophages are now considered as pro- sion number SNU-130311-2-2). Bone marrow derived macrophages inflammatory macrophages that display a robust immune re- (BMMs) were isolated as described before (16). To enrich the macro- sponses upon LPS stimulation compared with M-CSF grown phage population, we supplemented complete medium with 25 ng/ml macrophages (6). In contrast, M-CSF contributes the mainte- murine GM-CSF (Miltenyi Biotech, Bergisch Gladbach, Germany) for nance of most resident macrophages including osteoclast in GM-CSF grown bone marrow derived macrophages (GM-BMMs) or vivo and is known to affect homeostatic anti-inflammatory 20% L929 murine fibrosarcoma cell line culture supernatants for M-CSF grown bone marrow derived macrophage (M-BMMs). After 7 characteristics of macrophages. M-CSF grown macrophages days of differentiation, GM-BMMs and M-BMMs were stimulated with are widely accepted as in vitro-generated macrophage lipopolysaccharide (E. coli LPS, Sigma) 100 ng/ml for 2 h. Four ex- sources because they showed relatively homogenous and perimental groups (GM-BMM, GM-BMM/LPS, M-BMM, M-BMM/ stable macrophage phenotypes (7). A number of genomic LPS) were processed for further analysis. studies have been performed to analyze macrophage activa- Flow Cytometry—Macrophages were scrapped and incubated for 20 min with appropriate antibodies diluted to optimal concentrations tion in response to pro-inflammatory/anti-inflammatory stim- in FACS buffer (PBS, 5% FBS, 5 mM EDTA, and 1% NaN3). For cell uli. However, to date, there have been no clear reports on the sorting and proteomics analysis, antibodies included anti-CD45 (30- global proteomic differences that govern the functional char- F11), F4/80 (BM8), CD11b (M1/70), and IA/IE (M5/114.15.2). For sur- acteristics of differently differentiated or activated macro- face marker expression analysis, additional antibodies included phages (8–11). To fully elucidate what enables pro-inflamma- CD11c (HL3), CD80 (L307.4), MerTK (clone 125518, R&D systems, tory macrophages to be poised for rapid and robust immune Minneapolis, MN), and CD64 (X54–5/7.1, BD Biosciences, Franklin Lakes, NJ) and all were purchased from eBioscience unless indicated. responses requires assessing their global intracellular pro- Cells were sorted on a FACSAria or data were acquired on a LSRII teomic network signatures. flow cytometer (BD Biosciences). The surface marker expressions There has long been an appreciation, especially in the were analyzed using the FlowJo software. cancer field, for how changes in cellular activation coincide Trypsin Digestion, TMT Labeling, and OFF-Gel Fractionation— with alterations in cellular metabolic states (12, 13). Impor- Equal protein amounts were digested using Filter Aided Sample Prep- aration (FASP) method (17). Briefly, lysates were washed with 8 M tantly, over the last couple of years it is becoming increas- urea, followed by alkylation with 50 mM iodoacetamide (20 min at RT). ingly clear that immune cell activation is also coupled to After alkylation, filters were washed with 8 M Urea two times and with profound changes in cellular metabolism and that their fate 0.1 M TEAB two times for labeling with TMT reagents. Trypsin was and function are metabolically regulated (14). In line with added at a ratio of 1 g trypsin : 50 g protein and samples were this, in this study, we found that GM-CSF grown macro- incubated overnight at 37 °C. Tryptic digests of the four cell lysates phages have a higher glycolytic capacity through up-regu- were labeled with four different mass-tags among the TMT sixplex reagents respectively, per the protocol described by Thermo Fisher lated glycolytic enzymes, as well as high lipid/nitrogen com- Scientific (Rockford, IL). Resulting TMT-labeled peptides were subject pound biosynthetic enzymes compared with M-CSF grown to peptide isoelectrofocusing (IEF) fractionation, the 3100 OFFGEL macrophages. They produce robust inflammatory cytokines fractionator with a “Low Resolution Kit” pH 3–10 (Agilent Technolo- upon TLR ligand stimulation only when sufficient glucose is gies, Santa Clara, CA) according to the manufacturer’s instructions. available. Phosphopeptide Enrichment Using TiO —Phosphopeptide enrich- Tm ment was carried out as described in the Titansphere Phos-TiO Kit Here we performed a quantitative analysis of the pro- 2 manual. Briefly, phosphopeptides were eluted with 5% aqueous am- teome/phosphoproteome of primary GM-CSF and M-CSF monium hydroxide and 5% aqueous pyrrolidine solutions from Phos- grown macrophages using the latest isobaric tag based TiO (3 mg/200 l, Titansphere, GL Sciences Inc, Tokyo, Japan) spin TMT labeling and LC-MS/MS (15). This proteomic approach column. with high throughput technology is the first attempt to show Mass Spectrometric Analysis and Database Search—The extracted tryptic peptides were analyzed using a Q-Exactive mass spectrome- the fundamental differences between primary GM-CSF and ter (Thermo Fisher Scientific, Bremen, Germany) coupled with an M-CSF grown macrophages and finally reveals that innate Easy-nLC system (Thermo Fisher Scientific, Odense, Denmark). Tryp- cellular anabolic metabolism paves the way for inducing tic peptides were resuspended in 0.1% formic acid and separated on robust immune responses. In this study, we describe indi- EASY-Spray column (C18, 2 m particle size, 100 Å pore size, 75 m vidual differentially expressed proteins in the total network id  50 cm length, Thermo Fisher Scientific). Samples were resolved maps of GM-CSF and M-CSF grown macrophages and with a linear gradient of solvent B (100% ACN, 0.1% formic acid); 5–50% over 76 min, 50–90% over 12 min at a flow rate of 300 nL/min. predict how they are specifically involved in initiating inflam- The separated peptide ions eluted from the analytic column were mation or resolution. entered into the mass spectrometer at an electrospray voltage of 2.1 kV. All MS/MS spectra were acquired in a data-dependent mode for EXPERIMENTAL PROCEDURES fragmentation of the ten most abundant peaks from the full MS scan Macrophages Preparation—C57BL/6J mice were obtained from with 30% normalized collision energy. The dynamic exclusion dura- Jackson Laboratory (Bar Harbor, ME). Male mice of five to 10 weeks tion was set at 20 s and the isolation mass width was 2.5 Da. MS of age were used to isolate bone marrow cells. The mice were housed spectra were acquired with a mass range of 400–1800m/z and 70,000 under specific pathogen-free (SPF) conditions and cared according to resolution at m/z 200. MS/MS resolution was acquired at a resolution the Guide for the Care and Use of Laboratory Animals prepared by the of 17,500. The acquired MS/MS spectra were searched against the Institution of Animal Care and Use Committee of Seoul National Universal Protein Resource mouse protein database (Uniprot release University. All of the experiments were approved by the Institute for 2013_09, 50287 entries, http://www.uniprot.org/) with the Sequest Molecular & Cellular Proteomics 14.10 2723 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages algorithm in Proteome Discoverer 1.4 (Thermo Fisher Scientific, Bre- Seahorse cell plates (5  10 cells per well). Before plate reading, men, Germany). Search parameters were as follows: tryptic specificity cells were washed three times with glucose free assay media (Sea- with up to two missed cleavage sites, mass tolerances for precursor horse Bioscience), and the OCR and ECAR were assessed in glu- ions and fragment ions were set to 10 ppm and 0.8 Da, respectively, cose-containing assay media. Perturbation profiling of GM-BMMs fixed modification for carbamidomethyl-cysteine, TMT 4-plex of ly- and M-BMMs metabolic pathways was achieved by the addition of M), oligomycin (5 M), 2-deoxyglucose (100 mM) or LPS. sine and N terminus and variable modification for methionine oxida- glucose (10 m Experiments with the Seahorse system were done with the following tion. Following database searching, the output files were imported assay conditions: 2 min mixture; 2 min wait; and 4–5 min measure- into Scaffold Q (version Scaffold_4.3.2, Proteome Software Inc. ment. Metabolic parameters were then calculated. Portland, OR). Scaffold was used to organize all data, to quantitate Statistical Analysis—All data unless otherwise indicated are shown protein and to validate peptide identifications using the Peptide as mean  S.E. and were tested using two-tailed Student’s t test or Prophet algorithm (18). We selected peptides with cutoff of FDR two-way ANOVA using GraphPad Prism 4. 0.01 and Peptide threshold 95%. We then identified the proteins that have Protein threshold 99.9%. Validated data were normalized be- RESULTS tween intrasample channels and TMT signals showed at least 1.5-fold change in abundance were used for quantitative analysis of protein. Quantitative Analysis of Proteome and Phosphoproteome of The mass spectrometry proteomics data have been deposited to the GM-BMMs and M-BMMs—Macrophage sorting for proteomic ProteomeXchange Consortium (19) via the PRIDE partner repository analysis was performed according to the gating strategy with the dataset identifier PXD002582. shown in Fig. 1A. We first characterized in vitro GM-CSF GO Analyses—Gene Ontology biological processes (GOBPs) rep- resented by the sets of proteins were determined using the DAVID grown bone marrow derived macrophages (GM-BMMs) and software (20). For each set of proteins, we identified the GOBPs M-CSF grown bone marrow derived macrophages (M- represented by the genes with p value  0.05 using STRING v9.1 (21). BMMs). GM-GMMs were CD45 , F4/80 , and MHCII .M- The degree of centrality (K) and shortest-path centrality (SP) were BMMs showed homogenous population and expressed computed using CentiScaPe (22), and the nonseed proteins with CD45, F4/80, and CD11b. GM-BMMs expressed CD11c, SP  0orK  1 were removed. The nodes with the same GOBPs were grouped into the same modules, each of which was named by CD80 as well as the macrophage marker CD64 (Fig. 1B)(26). the corresponding GOBP. The network was visualized using Cyto- In contrast, M-BMMs expressed MHCII, CD11c, and CD80 scape (v. 2.8.3) (23). scarcely, indicating poor antigen presentation ability. Instead, RNA Isolation and Gene Expression Profiling—Global gene expres- they expressed relatively higher MerTK, which was further sion analyses were performed using Affymetrix GeneChip® Mouse up-regulated by M-CSF (27). Further experiments used sorted Gene 1.0 ST oligonucleotide arrays. The sample preparation was GM-BMMs and M-BMMs according to the above gating strat- performed according to the instructions and recommendations pro- vided by the manufacturer. Total RNA was extracted by Trizol reagent egy. In case of GM-CSF cultured cells, small population of (Invitrogen, Carlsbad, CA) according to the manufacturer’s instruc-  low high CD45 F4/80 MHCII dendritic cells were excluded tions. Expression data were generated by Affymetrix Expression Con- throughout the experiments. To enrich early TLR-responsive sole software version1.1. For normalization, RMA (Robust Multi-Av- proteins as well as core polarization specific proteins, we had erage) algorithm was implemented in the Affymetrix Expression four experimental groups of GM-BMMs and M-BMMs, with Console software. Western blotting—Sorted GM-BMMs and M-BMMs were lysed and and without LPS stimulation for 2 h (Fig. 1C). Three biologic analyzed for protein expression as described earlier (24). Antilactate replicates were generated by combining the protein extracts dehydrogenase B chain (Thermo), antiphosphofructokinse-1 (Santa- from five mice per group. The TMT labeling (fourplex) quanti- Cruz, TX), antiribose phosphate isomerase a (Abcam, Cambridge, tative proteomic method was used to profile changes in both UK), antitransferrin receptor (SantaCruz) and antibeta actin (Santa- protein abundance and phosphorylation stoichiometry in the Cruz) were used at 1/2000 dilutions. Cytokine Production—Sorted GM-BMMs and M-BMMs were whole cell extracts. Comparative MS analysis of these ex- seeded in 96-well plates at 5  10 /200 l with complete media and tracts on a Q-Exactive mass spectrometer led to the identifi- rested for 6 h. Macrophages were stimulated with LPS (100 ng/ml) for cation of 33,472 unique peptides from 3990 proteins with an 4 h and 24 h. Supernatants were collected and stored at 80 °C until average sequence coverage of 23% (Fig. 1D, supplemental TNF, IL-6, and IL-10 quantification. For IL-1 quantitation, cells were table S12-S13). Although expression levels remained unaf- lysed in RIPA buffer and stored at 80 °C. Cells in each well were fected by TLR stimulation for most of the identified proteins of analyzed for their protein concentration using the BCA assay kit (Thermo, Waltham, MA) and this was used as a normalization factor GM-BMMs and M-BMMs, a subset of 294 proteins showed for cytokine quantitation. Cytokine concentrations were determined more than a 1.5-fold change in abundance (supplemental using the duoset ELISA kit (R&D systems, Minneapolis, MN). Tables S1 and S2). Latex Bead Phagocytosis—Sorted GM-BMMs and M-BMMs were 6 We next examined the differences in the status of protein seeded in 12-well plates at 10 /ml with complete media and rested for phosphorylation between GM-BMMs and M-BMMs as well as 6 h. Each well of cells were incubated with 10 Alexa 350-tagged latex beads (Molecular Probe, Eugene, OR) of 1 m diameter for 2 h. The GM-BMMs/LPS and M-BMMs/LPS (Fig. 1E). This analysis led number of phagocytozed beads were analyzed on a LSRII flow cy- to the identification of 2239 unique phosphopeptides from tometer (BD Biosciences) and analyzed using the FlowJo software. 1274 phosphoproteins (4844 phosphorylation sites, peptide ECAR and OCR Measurements—The Seahorse XF-24 metabolic probability95% and protein probability99%). A compari- extracellular flux analyzer (Seahorse Bioscience, Billerica, MA) was son of the identified phosphorylation sites with those from used to analyze the ECAR (in mpH/min) and the OCR (in pmol/min) PHOSIDA (28), PhosphoSitePlus (29), and PhosphoELM (30) (25). Briefly, GM-BMMs and M-BMMs were differentiated for 7 days in 2724 Molecular & Cellular Proteomics 14.10 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages FIG.1. Quantitative proteomic/phosphoproteomic analysis of GM-BMMs/M-BMMs. A, GM-BMMs and M-BMMs were differentiated over 7 days from C57BL/6J male mice bone marrow cells with GM-CSF or M-CSF supplementation, respectively. Cells were sorted by CD45 F4/ 80 MHCII population for GM-BMMs, or CD45 F4/80 CD11b population for M-BMMs. B, Surface marker examination of GM-BMMs and M-BMMs. Attached cells were stained with the appropriate antibodies and analyzed for surface MHCII, CD11c, CD80, CD64, and MerTK expressions per gated populations in A. Three independent experiments were performed. **, p  0.01, ***, p  0.001 by Student t test. C, Experimental strategy. Macrophage proteins were pooled to generate three biological replicates for each of the GM-BMM/M-BMM/GM- BMMLPS/M-BMMLPS experimental groups. D, Results of protein expression experiments. E, Results of phosphoproteome experiments. F, Scatter plot of gene symbols detected both in proteome and microarray analyses. Data represents the fold change of log (GM-BMM/M-BMM) of 2491 genes. Regression p value  2.1987e-242, R-squre  0.31352. G, GO analysis of differentially expressed proteins and phosphopeptides. Heat-map shows significant GO biological process terms (p  0.05) for differentially expressed proteins/phosphoproteins between GM-BMM/M- BMM and GM-BMMLPS/M-BMMLPS. Red color indicates increased protein/phosphoprotein abundances in GM-BMM or GM-BMM/LPS, and blue color indicates increased abundances in M-BMM or M-BMM/LPS. Star markings show the most enriched GO category in a heatmap. Molecular & Cellular Proteomics 14.10 2725 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages TABLE I List of proteins enriched in Gene Ontology Biological Process category from differentially expressed protein abundances between GM-BMMs versus M-BMMs and GM-BMMsLPS vs M-BMMsLPS: upregulated in GM-BMMs Gene symbol Protein Name Ratio GM-BMM/M-BMM p value Fold enrichment Glucose metabolic process 0.00006 7.84 ALDOC Fructose-bisphosphate aldolase C 2.22 ENO1 Alpha-enolase 1.57 FABP5 Fatty acid-binding protein, epidermal 2.64 GYS1 Glycogen starch synthase, muscle 1.68 LDHB L-lactate dehydrogenase B chain 2.00 PFKP 6-phosphofructokinase 1.80 PYGL Glycogen phosphorylase, liver form 1.62 RPIA Ribose-5-phosphate isomerase 1.62 Nitrogen compound biosynthetic process 0.02197 3.18 ASS1 Argininosuccinate synthetase 1 3.03 ATP6V0D2 V-type proton ATPase subunit d2 2.93 KYNU Kynureninase 1.62 PADI4 Protein-arginine deiminase type-4 1.74 PHGDH D-3-phosphoglycerate dehydrogenase 2.00 RRM1 Ribonucleoside-diphosphate reductase large subunit 1.93 SLC7A2 Low affinity cationic amino acid transporter 2 1.68 Lipid biosynthetic process 0.00021 4.80 ACLY ATP-citrate synthase 1.62 ALOX5 Arachidonate 5-lipoxygenase 1.68 CD74 H-2 class II histocompatibility antigen gamma chain 2.46 FABP5 Fatty acid-binding protein, epidermal 2.64 FASN Fatty acid synthase 1.62 FDPS Farnesyl pyrophosphate synthase 2.00 HMGCS1 Hydroxymethylglutaryl-CoA synthase, cytoplasmic 1.68 IDI1 Isopentenyl-diphosphate Delta-isomerase 1 1.93 LTA4H Leukotriene A4 hydrolase 2.00 MVD Diphosphomevalonate decarboxylase 2.00 indicated that 2541 sites ( 52%) have not been previously differentially phosphorylated proteins (DPPs) in GM-BMMs/ reported. We also evaluated the confidence level of site lo- M-BMMs and GM-BMMs/LPS M-BMMs/LPS pairs (Fig. 1G). calization using a probability score function and determined A total of 52 categories were enriched and are presented in a that 34% of the identified sites (1635 sites) corresponded to heatmap. Interestingly, the most obvious enriched categories high confidence assignments (supplemental Table S11). in both GM-BMMs and GM-BMMs/LPS macrophages were Among them, 450 and 899 phosphopeptides, corresponding metabolic processes (Table I). These include carbohydrate to 228 and 481 proteins, were differentially regulated between catabolic, glucose metabolic, alcohol catabolic, lipid biosyn- GM-BMMs and M-BMMs upon LPS stimulation, respectively thetic, and amino acid metabolic processes (Table I). The GO (supplemental Table S8). term macromolecular complex subunit organization was en- A scatter plot of protein abundance and mRNA expression hanced in GM-BMMs/LPS compared with GM-BMMs. These (GEO accession number GSE63245) showed a general pre- energy consuming processes are potentially supported by diction ability of microarray data to project protein quantity glucose metabolic/nitrogen compound biosynthetic and lipid (Fig. 1F), but is still limited in representing exact whole protein synthetic pathways in GM-BMMs, as with rapidly proliferating quantities of macrophages with a regression p value of cancer cells (31). Secretion, chromatin organization, and apo- 2.1987e ptosis were specifically up-regulated in GM-BMMs/LPS. and R-square of 0.31352. This means that sub- stantial macrophage proteins go through post-translational In contrast, M-BMMs and M-BMMs/LPS had different en- modification and our proteomics data could give valuable and riched functions, in that they showed higher endocytosis and precise insight into these global differences between opposite homeostasis processes. Of note, “energy derivation by oxi- characterized macrophages. dation of organic compounds” was highly up-regulated in GO Analysis Revealed Metabolism as a Key Regulator of M-BMMs (Fig. 1G), showing a metabolic state that is the Macrophage Function—To gain insight into the functional dif- converse of GM-BMMs (Table II). This decreased upon LPS ferences of each macrophage type, we conducted a compar- stimulation, implying a metabolic shift from oxidation to other ison of Gene Ontology (GO) annotations of biological pro- processes. Subcellular distribution of DEPs also showed that cesses from differentially expressed proteins (DEPs) and mitochondrial components were highly enriched in M-BMMs 2726 Molecular & Cellular Proteomics 14.10 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages TABLE II List of proteins enriched in Gene Ontology Biological Process category from differentially expressed protein abundances between GM-BMMs versus M-BMMs and GM-BMMsLPS vs M-BMMsLPS: upregulated in M-BMMs Fold Gene symbol Protein Name Ratio M-BMM/GM-BMM p value enrichment Energy derivation by oxidation of organic compounds 0.00180 6.82 CAT Catalase 1.80 GAA Lysosomal alpha-glucosidase 1.57 SDHC Succinate dehydrogenase cytochrome b560 subunit, 1.74 mitochondrial SLC37A2 Sugar phosphate exchanger 2 1.74 SOD2 Superoxide dismutase Mn , mitochondrial 2.46 SUCLG2 Succinyl-CoA ligase GDP-forming subunit beta, 1.57 mitochondrial Endocytosis 0.00004 5.9 ABCA1 ATP-binding cassette, sub-family A (ABC1), member 1 1.57 ARHGAP27 Rho GTPase activating protein 27 1.57 BET1 Bet1 golgi vesicular membrane trafficking protein 1.52 CD36 CD36 molecule (thrombospondin receptor) 1.80 EHD1 EH-domain containing 1 2.22 ELMO1 engulfment and cell motility 1 1.52 FNBP1L formin binding protein 1-like 2.00 HCK hemopoietic cell kinase 1.74 ITSN1 intersectin 1 (SH3 domain protein) 2.14 TFRC transferrin receptor 2.46 VAV1 vav 1 guanine nucleotide exchange factor 1.57 compared to GM-BMMs, suggesting a bias for oxidation pro- that is indirectly related to glycolysis (Fig. 2A). Among these, cesses as their cellular energy source (supplemental Table we confirmed the abundant protein expressions of ribose 5 S9). Several signaling pathways related to endocytosis/phag- phosphate isomerase (RPIA) and 6-phosphofructokinase ocytosis, including small GTPase and Ras, were enriched in (PFKP) in GM-BMMs compared with M-BMMs by Western both M-BMMs and M-BMMs/LPS. blot, indicating the reliability of our proteomic methodology Similar results were also obtained in our integrated pathway (Fig. 2B). analysis (IPA) of DPPs between GM-BMMs and GM-BMMs/ Next we hypothesized that GM-BMMs have a much higher LPS or M-BMMs and M-BMMs/LPS. GM-BMMs/LPS showed maximum glycolytic capacity than M-BMMs. To assess met- enhanced protein phosphorylations for cytokine signaling, abolic differences, we recorded extracellular acidification rate MAPK signaling, TLR4 cascade as well as mTOR signaling (ECAR) and oxygen consumption rate (OCR) of GM-BMMs events compared with basal GM-BMMs (supplemental Table and M-BMMs. GM-BMMs had significantly higher basal S4). In line with this, 58 proteins were up-regulated in GM- ECAR than M-BMMs, although OCR was similar (Fig. 2C). BMMs/LPS compared with GM-BMMs, mostly related to im- Next we treated the macrophages with glucose and oligomy- mune responses including IL-1, lactotransferrin, TNF, cop- cin to maximize glycolysis as shown in Fig. 2D. This con- per transporter protein ATOX-1, and IL-1 (supplemental firmed that GM-BMMs have a higher glycolytic capacity Table S3). M-BMMs/LPS showed enhanced Rho GTPase and than M-BMMs (Fig. 2E). This difference contributes to the RAC1 activity in IPA of DPPs, which correlate with endocytic varied extent of acute glycolysis upon LPS stimulation, with activity. M-BMMs/LPS showed only seven up-regulated pro- GM-BMMs showing a more prominent and prolonged ECAR teins relative to basal M-BMMs (supplemental Table S7), sug- curve than M-BMMs after LPS treatment (Fig. 2F). Acute gesting remarkable down-regulated immune responses com- glycolysis in macrophages following LPS was totally de- pared with GM-BMMs. Taken together, GM-BMMs had pendent on glucose uptake as ECAR does not increase enhanced anabolic metabolism and responded to LPS with without glucose in the media (Fig. 2G). Taken together, inflammatory activities. On the contrary, M-BMMs had en- these results indicate that GM-CSF grown macrophages hanced endocytic processes and did not show marked in- have inherently higher glycolytic capacities than M-CSF flammatory activities upon LPS stimulation. grown macrophages to enable heightened glucose conver- Glycolysis Signatures GM-CSF Grown Macrophages— sion to lactate in response to LPS. Among the eight proteins involved in glucose metabolism Metabolic Differences are Linked to Cytokine Responses in up-regulated in GM-BMMs, five were direct glycolytic en- Macrophages—Finally, to test the impact of metabolic con- zymes (RPIA, PFKP, LDHB, ENO1, ALDOC) and one, PYGL, struct of GM-BMMs and M-BMMs on cytokine production, Molecular & Cellular Proteomics 14.10 2727 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages FIG.2. GM-BMMs have higher glycolytic capacities than M-BMMs. A, Expression profiles of glycolytic genes and proteins in GM-BMMs compared with M-BMMs. ***, p  0.001 by two-way ANOVA. B, Western blot results for RPIA, PFKP, and actin. C, Basal OCR and ECAR of GM-BMMs and M-BMMs. Data obtained using XF-24 extracellular flux analyzer. Three independent experiments were performed. **, p  0.01 by Student t test. D, Maximum glycolytic capacities of GM-BMMs and M-BMMs. Perturbation profiling of lactate production (ECAR) was achieved by the addition of glucose (10 mM), oligomycin (5 M), and 2-deoxyglucose (2-DG, 100 mM). Three independent experiments were performed. ***, p  0.001 by two-way ANOVA. E, Quantitative graph representing mean maximum glycolytic capacities of GM-BMMs and M-BMMs. Three independent experiments were performed. *, p  0.05 by Student t test. F, LPS induced acute glycolysis. GM-BMMs and M-BMMs were stimulated with 100 ng/ml E. coli LPS and recorded for the ECAR changes. Data showing relative ECAR (%) changes from baseline. Three independent experiments were performed. ***, p  0.001 by two-way ANOVA. G, Glucose effect on acute glycolysis upon LPS stimulation. GM-BMMs were stimulated with LPS with or without media glucose and ECAR changes were recorded. H–K, GM-BMMs produced more inflammatory cytokines than M-BMMs in a glycolysis-dependent manner. GM-BMMs and M-BMMs were sorted and replated in 96-well plates at 5  10 density. Cells were treated with LPS (100 ng/ml) with or without 2-DG (100 mM) and cytokines were measured at 4 h (TNF) and 24 h (IL-6, IL-1, IL-10). ELISA was performed to quantify TNF (H), IL-6 (I), IL-10 (K) production in culture supernatants, or IL-1 synthesis (J) in cell lysates. Three independent experiments were performed. **, p  0.01, ***, p  0.001 by two-way ANOVA. we employed 2-deoxyglucose (2-DG) to interfere with glu- upregulated endocytosis-relating proteins in M-BMMs cose metabolism and measured cytokine production as a (supplemental Table S5). Among the upregulated 11 proteins functional readout. Consistent with previous reports (6), in M-BMMs compared to GM-BMMs, we confirmed that GM-BMMs produced more TNF, IL-6, and IL-1 than M- transferrin receptor, CD71 (TFRC), which transports trans- BMMs (Fig. 2H,2I, and 2J). Importantly, blockade of glycol- ferrin inside the cell and is up-regulated by CSF1 (32), ysis significantly reduced LPS-induced TNF, IL-6, and expression was higher in M-BMMs than GM-BMMs (Fig. IL-1 production to that of basal levels both in GM-BMMs 3B). M-BMMs showed enhanced endocytic processes both and M-BMMs. Both macrophages produced similar levels of in protein expression as well as phosphoprotein enrichment. IL-10 upon LPS stimulation, but this was also inhibited by Thus, we performed latex bead uptake experiments (Fig. 2-DG (Fig. 2K). These results confirm the vital role of gly- 3C). After GM-BMMs and M-BMMs sorting according to colysis in LPS induced pro-inflammatory cytokine re- Fig. 1A, replated macrophages were incubated with Ax350- sponses. Given that LPS induced cytokine production is tagged latex beads of 1 m diameter for 2 h. As expected largely dependent on glycolysis, it goes to reason that the 40% of M-BMMs had more than four beads per cell, but more glycolytically active GM-BMMs can synthesize con- only 20% of GM-BMMs did (Fig. 3D). These results con- siderably more inflammatory cytokines than homeostatic firmed the relationship between our proteome/phosphopro- M-BMMs. teome analysis and macrophage functional characteristics Endocytosis Predominates in M-CSF Grown Macrophag- described above. es—Eleven proteins related to endocytosis were up-regulated Protein Interaction Network Analysis of Macrophages—To in M-BMMs compared to GM-BMMs (Fig. 3A). LPS further gain insight into the proteome and phosphoproteome of GM- 2728 Molecular & Cellular Proteomics 14.10 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages FIG.3. M-BMMs phagocytosed more beads than GM-BMMs. A, Relative mRNA and protein expression levels of endocytosis participants in M-BMMs compared with GM-BMMs. **, p  0.01, ***, p  0.001 by two-way ANOVA. B, Western blot results for TFRC and actin. C, Histograms showing the number of beads per cell. Sorted macrophages were replated and incubated with Alexa350 tagged latex beads for 2 h. Bead phagocytosis was analyzed per macrophage population (GM-BMMs: F4/80 MHCII , M-BMMs: F4/80 CD11b ) using FACS. Cell percentages containing more than four beads are indicated in both histograms. D, Quantitative graph showing cell percentages with more than four beads per cells. Three independent experiments were performed. *, p  0.05 by Student t test. BMMs and M-BMMs, we conducted an Ingenuity Pathway replication. In M-CSF grown macrophages, Rho/GTPase sig- Analysis (IPA) using DEPs and DPPs involved in differentially naling is well interconnected with endocytosis and adhesion. regulated biological processes of six major groups between DISCUSSION GM-BMMs and M-BMMs: (1) Glucose metabolism, (2) Lipid metabolism, (3) Amino acid metabolism, (4) DNA replication, In this study, we used the latest isobaric tag based mul- (5) Endocytosis, and (6) Rho/GTPase signaling. The total net- tiplex proteomic/phosphoproteomic quantitative analysis work map is presented in Fig. 4, which depicts 145 DEPs or method to identify key molecular differences between GM- DPPs in colored figures as well as predicted regulating tran- CSF and M-CSF grown macrophages. We found that GM- scription factors, transporters, enzymes, and signaling mole- BMMs have potentiative glycolytic/lipid biosynthetic path- cules denoted with white circles. Of note, transcription factors ways as well as nitrogen compound biosynthesis processes. that regulate glycolysis such as HIF1A, MYC, FOXO1, FOXA2, These enhanced anabolic pathways directly link to their pro- ATF4, HNF1A, and HNF1B, underpin various other processes inflammatory cytokine production capacity. We also pre- in GM-CSF grown macrophages including glucose/lipid/ sented valuable targets for disrupting macrophage functions, amino acid metabolism, immune response as well as DNA as well as novel phosphosites of macrophage proteins Molecular & Cellular Proteomics 14.10 2729 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages FIG.4. Regulated protein networks in GM-BMMs versus M-BMMs. Protein abundance and phosphorylation data were analyzed with the Ingenuity Pathway Analysis (IPA) using 145 differentially expressed proteins and phosphoproteins involved in differentially regulated biological process between GM-BMMs and M-BMMs. Glucose metabolism, lipid metabolism, amino acid metabolism, DNA replication, endocytosis, and Rho/GTPase signaling were the subnetworks most significantly affected. Proteins/nodes are grouped according to their function. Red color indicates GM-BMMs up-regulation and green is M-BMMs up-regulation. Circle means DEP, diamond means DPP. White circles are predicted interacting signaling molecules as well as transcription factors. All indicated by gene symbol. (supplemental Table S11). To our knowledge, this is the first Of note, the role of glycolysis in GM-BMMs differs from the large scale proteome/phosphoproteome analysis of GM-CSF acute glycolytic switch following LPS stimulation in dendritic and M-CSF grown macrophages. Our reliable data gives in- cells, which results from IKK/TBK1 mediated Akt activation sight into the importance of cellular metabolic polarization and (37), because basal enhanced glycolytic state exists in GM- their functional outcomes in macrophages. CSF primed macrophages in the absence of TLR signaling. It Aberrant carbohydrate metabolism is classically a distinc- also appears that this bias toward glycolysis does not stem tive feature of tumor cells (33). In general, normal cells pro- from TLR mediated inducible nitric oxide synthase (iNOS) duce most of their ATP from glucose through oxidative phos- expression, which produces nitric oxide (NO) and inhibit oxi- phorylation. However, many cancer cells produce ATP by dative phosphorylation (38). Instead, our network analysis conversion of glucose to lactate and therefore exhibit lower predicts that GM-BMMs have up-regulated upstream glucose oxidative phosphorylation activity. This “glycolytic pheno- regulatory factors such as HIF1A, MYC, HNF1A (39), or type” ensures sufficient macromolecule biosynthesis neces- FOXA2 (40). Given that GM-CSF is known to increase L-Myc sary for rapid cell growth and division (34). Compellingly, we expression in dendritic cells after 24 h (41), glycolysis in found that GM-BMMs showed a metabolic state similar to GM-CSF grown macrophages may be possibly further up- cancer cells. We confirmed that the up-regulated glycolytic regulated by Myc because many glycolytic enzymes have enzymes are collectively involved in LPS-induced glycolysis myc-binding sites in their promoters (42). In 1995, Brissette et and consequently connected with inflammatory cytokine pro- al. reported that GM-CSF primes mice for enhanced cytokine ductions. These data give us profound implications in that production in response to LPS, and our results suggest a GM-CSF affected macrophages are usually generated when possible underlying mechanism (43). our body has been influenced by microbial or aseptic inflam- Another interesting result was that GM-BMMs have up- matory stimuli (35). Given that GM-CSF therapy has been regulated enzymes in the mevalonate pathway. They are in- considered to alleviate inflammatory conditions such as ar- cluded in the GO term lipid biosynthetic process: diphospho- thritis (36), enhanced glycolytic pathway of GM-CSF grown mevalonate decarboxylase (MVD), farnesyl pyrophosphate macrophages might be an alternative specific valuable target synthase (FDPS), hydroxymethylglutaryl-CoA synthase for macrophage mediated inflammatory diseases. (HMGCS1), and isopentenyl-diphosphate delta-isomerase 1 2730 Molecular & Cellular Proteomics 14.10 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages (IDI1) (Table I). Similar to glycolysis, the mevalonate pathway granulocyte-macrophage colony-stimulating factor. Crit. Rev. Immunol. 25, 405–428 is dysregulated in tumor cells (44) and several researchers 5. Hamilton, J. A. (2008) Colony-stimulating factors in inflammation and au- have tried to target this pathway to inhibit cancer cell malig- toimmunity. Nat. Rev. Immunol. 8, 533–544 nancy (45). The mevalonate pathway is a complex biochem- 6. Fleetwood, A. J., Lawrence, T., Hamilton, J. A., and Cook, A. D. (2007) Granulocyte-macrophage colony-stimulating factor (CSF) and macro- ical pathway that generates several fundamental end-prod- phage CSF-dependent macrophage phenotypes display differences in ucts including cholesterol, isoprenoids, dolichol, ubiquinone, cytokine profiles and transcription factor activities: implications for CSF and isopentenyladenine (45). In fact, glycolysis cascades into blockade in inflammation. J. Immunol. 178, 5245–5252 7. Murray, P. J., Allen, J. E., Biswas, S. K., Fisher, E. A., Gilroy, D. W., Goerdt, the mevalonate pathway when macrophages are activated S., Gordon, S., Hamilton, J. A., Ivashkiv, L. B., Lawrence, T., Locati, M., because glycolysis induced by TLR activation produces Mantovani, A., Martinez, F. O., Mege, J. L., Mosser, D. M., Natoli, G., abundant pyruvate and acetyl-CoA followed by lipid accumu- Saeij, J. P., Schultze, J. L., Shirey, K. A., Sica, A., Suttles, J., Udalova, I., van Ginderachter, J. A., Vogel, S. N., and Wynn, T. A. (2014) Macrophage lation inside cells (46). Thus, it would be interesting to inves- activation and polarization: nomenclature and experimental guidelines. tigate whether the mevalonate pathway contributes to the Immunity 41, 14–20 macrophage pro-inflammatory phenotype. 8. Martinez, F. O., Gordon, S., Locati, M., and Mantovani, A. (2006) Transcrip- tional profiling of the human monocyte-to-macrophage differentiation The immune system is comprised of a heterogeneous pop- and polarization: new molecules and patterns of gene expression. J. Im- ulation of cells that, for the most part, are relatively quiescent munol. 177, 7303–7311 in the steady state but share the ability to rapidly respond to 9. McDermott, J. E., Archuleta, M., Thrall, B. D., Adkins, J. N., and Waters, K. M. (2011) Controlling the response: predictive modeling of a highly infection, inflammation, and other perturbations. There is a central, pathogen-targeted core response module in macrophage acti- growing appreciation of the fact that transitions between qui- vation. PloS One 6, e14673 escent and activated states require the apportioning of nutri- 10. Nau, G. J., Richmond, J. F., Schlesinger, A., Jennings, E. G., Lander, E. S., and Young, R. A. (2002) Human macrophage activation programs in- ents into different pathways, and, therefore, there is a strong duced by bacterial pathogens. Proc. Natl. Acad. Sci. U.S.A. 99, interest in how metabolic pathways are regulated to support 1503–1508 or direct functional changes. Several reports have been al- 11. Ramsey, S. A., Klemm, S. L., Zak, D. E., Kennedy, K. A., Thorsson, V., Li, ready noted the importance of macrophage metabolic states B., Gilchrist, M., Gold, E. S., Johnson, C. D., Litvak, V., Navarro, G., Roach, J. C., Rosenberger, C. M., Rust, A. G., Yudkovsky, N., Aderem, on polarization and attempted to target various metabolic A., and Shmulevich, I. (2008) Uncovering a macrophage transcriptional pathways, such as glucose-6-phosphate dehydrogenase (47), program by integrating evidence from motif scanning and expression CARKL (48), PGC-1 dynamics. PLoS Comput. Biol. 4, e1000021 (49), TRAP1 (50), or glucose transporter 12. Vander Heiden, M. G., Cantley, L. C., and Thompson, C. B. (2009) Under- 1(51), to modulate macrophage phenotype. Our data clearly standing the Warburg effect: the metabolic requirements of cell prolifer- reveals a systemic link between macrophage activation and ation. Science 324, 1029–1033 metabolic state. In addition, this study identifies novel oppor- 13. Raffaello, A. and Rizzuto, R. (2011) Mitochondrial longevity pathways. Biochim Biophys Acta. 1813, 260–268 tunities for targeting macrophage mediated immune re- 14. Pearce, E. L., and Pearce, E. J. (2013) Metabolic pathways in immune cell sponses. activation and quiescence. Immunity 38, 633–643 15. Dayon, L., and Sanchez, J. C. (2012) Relative protein quantification by * This work was supported by Basic Science Research Program MS/MS using the tandem mass tag technology. Methods Mol. Biol. 893, through the National Research Foundation of Korea (NRF) funded 115–127 by the Ministry of Science, ICT & Future Planning (NRF- 16. Na, Y. R., Yoon, Y. N., Son, D. I., and Seok, S. H. (2013) Cyclooxygenase-2 2014R1A1A1008012, NRF-2012R1A1A3013393) and supported by inhibition blocks M2 macrophage differentiation and suppresses metas- tasis in murine breast cancer model. PloS One 8, e63451 Proteogenomic Research Program, and the Bio- and Medical Tech- 17. AM, V. O. s., Dominin, S. G., Kutepov, E. N., Leonov, A. V., Lobov, A. V., nology Development Program (Project No. 2012M3A9B6055305) Maimulov, V. G., Nesvizhskii Iu, V., Semenova, V. V., Fokin, M. V., and through the National Research Foundation of Korea funded by the Tselykovskaia, N. (2003) [Training physicians in medical prevention spe- Korean Ministry of Education, Science and Technology (to K.P.K.). cialty: problems and prospects]. Gigiena i Sanitariia, 1, 13–15 □ S This article contains supplemental Tables S1 to S13. 18. Keller, A., Nesvizhskii, A. I., Kolker, E., and Aebersold, R. (2002) Empirical ** These authors contributed equally to this work. statistical model to estimate the accuracy of peptide identifications made To whom correspondence should be addressed: Kyung Hee by MS/MS and database search. Anal. Chem. 74, 5383–5392 University, Dept. of Applied Chemistry, Rm471-4 Engineering 19. Vizcaino, J. A., Deutsch, E. W., Wang, R., Csordas, A., Reisinger, F., Ríos, BLDG 1732 Deokyeong-daero, Giheung-gu, Yongin 446-701, Re- D., Dianes, J. A., Sun, Z., Farrah, T., Bandeira, N., Binz, P. A., Xenarios, public of Korea. Tel.: 82-31-201-3868; Fax: 82-31-201-2825; E- I., Eisenacher, M., Mayer, G., Gatto, L., Campos, A., Chalkley, R. J., Kraus, H. J., Albar, J. P., Martinez-Bartolome, S., Apweiler, R., Omenn, mail: [email protected]. Seoul National University College of Medi- G. S., Martens, L., Jones, A. R., and Hermjakob, H. (2014) ProteomeX- cine, Dept. of Microbiology and Immunology, and Institute of change provides globally coordinated proteomics data submission and endemic disease, Seoul 110-799, Republic of Korea. Tel.: 82-2- dissemination. Nat. Biotechnol. 32, 223–226 740-8302; Fax: 82-2-743-0881; E-mail: [email protected]. 20. Huang da, W., Sherman, B. T., and Lempicki, R. A. (2009) Systematic and integrative analysis of large gene lists using DAVID bioinformatics re- REFERENCES sources. Nat. Protoc. 4, 44–57 1. Gordon, S., and Taylor, P. R. (2005) Monocyte and macrophage heteroge- 21. Franceschini, A., Szklarczyk, D., Frankild, S., Kuhn, M., Simonovic, M., Roth, A., Lin, J., Minguez, P., Bork, P., von Mering, C., and Jensen, L. J. neity. Nat. Rev. Immunol. 5, 953–964 (2013) STRING v9.1: protein–protein interaction networks, with increased 2. Mosser, D. M., and Edwards, J. P. (2008) Exploring the full spectrum of coverage and integration. Nucleic Acids Res. 41, D808–815 macrophage activation. Nat. Rev. Immunol. 8, 958–969 22. Scardoni, G., Petterlini, M., and Laudanna, C. (2009) Analyzing biological 3. Gordon, S. (2003) Alternative activation of macrophages. Nat. Rev. Immu- network parameters with CentiScaPe. Bioinformatics 25, 2857–2859 nol. 3, 23–35 4. Fleetwood, A. J., Cook, A. D., and Hamilton, J. A. (2005) Functions of 23. Smoot, M. E., Ono, K., Ruscheinski, J., Wang, P. L., and Ideker, T. (2011) Molecular & Cellular Proteomics 14.10 2731 Proteomic Analysis Between GM-CSF and M-CSF Grown Macrophages Cytoscape 2.8: new features for data integration and network visualiza- 39. Wang, H., Antinozzi, P. A., Hagenfeldt, K. A., Maechler, P., and Wollheim, tion. Bioinformatics 27, 431–432 C. B. (2000) Molecular targets of a human HNF1 alpha mutation respon- 24. Son, D., Na, Y. R., Hwang, E. S., and Seok, S. H. (2014) Platelet-derived sible for pancreatic beta-cell dysfunction. EMBO J. 19, 4257–4264 growth factor-C (PDGF-C) induces anti-apoptotic effects on macro- 40. Wolfrum, C., Shih, D. Q., Kuwajima, S., Norris, A. W., Kahn, C. R., and phages through Akt and Bad phosphorylation. J. Biol. Chem. 289, Stoffel, M. (2003) Role of Foxa-2 in adipocyte metabolism and differen- 6225–6235 tiation. J. Clin. Invest. 112, 345–356 25. Spolarics, Z., and Wu, J. X. (1997) Role of glutathione and catalase in H2O2 41. Kc, W., Satpathy, A. T., Rapaport, A. S., Brisen˜ o, C. G., Wu, X., Albring, detoxification in LPS-activated hepatic endothelial and Kupffer cells. J. C., Russler-Germain, E. V., Kretzer, N. M., Durai, V., Persaud, S. P., Am. J. Physiol. 273, G1304–1311 Edelson, B. T., Loschko, J., Cella, M., Allen, P. M., Nussenzweig, M. C., 26. Gautier, E. L., Shay, T., Miller, J., Greter, M., Jakubzick, C., Ivanov, S., Helft, Colonna, M., Sleckman, B. P., Murphy, T. L., and Murphy, K. M. (2014) J., Chow, A., Elpek, K. G., Gordonov, S., Mazloom, A. R., Ma’ayan, A., L-Myc expression by dendritic cells is required for optimal T-cell priming. Chua, W. J., Hansen, T. H., Turley, S. J., Merad, M., Randolph, G. J., and Nature 507, 243–247 Immunological Genome, C. (2012) Gene-expression profiles and tran- 42. Gordan, J. D., Thompson, C. B., and Simon, M. C. (2007) HIF and c-Myc: scriptional regulatory pathways that underlie the identity and diversity of sibling rivals for control of cancer cell metabolism and proliferation. mouse tissue macrophages. Nat. Immunol. 13, 1118–1128 Cancer Cell 12, 108–113 27. Zizzo, G., Hilliard, B. A., Monestier, M., and Cohen, P. L. (2012) Efficient 43. Brissette, W. H., Baker, D. A., Stam, E. J., Umland, J. P., and Griffiths, R. J. clearance of early apoptotic cells by human macrophages requires M2c (1995) GM-CSF rapidly primes mice for enhanced cytokine production in polarization and MerTK induction. J. Immunol. 189, 3508–3520 response to LPS and TNF. Cytokine 7, 291–295 28. Gnad, F., Gunawardena, J., and Mann, M. (2011) PHOSIDA 2011: the 44. Clendening, J. W., Pandyra, A., Boutros, P. C., El Ghamrasni, S., Khosravi, post-translational modification database. Nucleic Acids Res. 39, F., Trentin, G. A., Martirosyan, A., Hakem, A., Hakem, R., Jurisica, I., and D253–260 Penn, L. Z. (2010) Dysregulation of the mevalonate pathway promotes 29. Hornbeck, P. V., Kornhauser, J. M., Tkachev, S., Zhang, B., Skrzypek, E., transformation. Proc. Natl. Acad. Sci. U.S.A. 107, 15051–15056 Murray, B., Latham, V., and Sullivan, M. (2012) PhosphoSitePlus: a 45. Swanson, K. M., and Hohl, R. J. (2006) Anti-cancer therapy: targeting the comprehensive resource for investigating the structure and function of mevalonate pathway. Curr. Cancer Drug Targets 6, 15–37 experimentally determined post-translational modifications in man and 46. Feingold, K. R., Shigenaga, J. K., Kazemi, M. R., McDonald, C. M., Patzek, mouse. Nucleic Acids Res. 40, D261–270 S. M., Cross, A. S., Moser, A., and Grunfeld, C. (2012) Mechanisms of 30. Dinkel, H., Chica, C., Via, A., Gould, C. M., Jensen, L. J., Gibson, T. J., and triglyceride accumulation in activated macrophages. J. Leukocyte Biol. Diella, F. (2011) Phospho.ELM: a database of phosphorylation 92, 829–839 sites–update 2011. Nucleic Acids Res. 39, D261–267 47. Ham, M., Lee, J. W., Choi, A. H., Jang, H., Choi, G., Park, J., Kozuka, C., 31. Imamura, K., Ogura, T., Kishimoto, A., Kaminishi, M., and Esumi, H. (2001) Sears, D. D., Masuzaki, H., and Kim, J. B. (2013) Macrophage glucose- Cell cycle regulation via p53 phosphorylation by a 5 -AMP activated 6-phosphate dehydrogenase stimulates proinflammatory responses with protein kinase activator, 5-aminoimidazole- 4-carboxamide-1-beta-D- oxidative stress. Mol. Cell. Biol. 33, 2425–2435 ribofuranoside, in a human hepatocellular carcinoma cell line. Biochem. 48. Haschemi, A., Kosma, P., Gille, L., Evans, C. R., Burant, C. F., Starkl, P., Biophys. Res. Commun. 287, 562–567 Knapp, B., Haas, R., Schmid, J. A., Jandl, C., Amir, S., Lubec, G., Park, 32. Lokeshwar, B. L., and Lin, H. S. (1990) Growth factor-dependent regulation J., Esterbauer, H., Bilban, M., Brizuela, L., Pospisilik, J. A., Otterbein, of transferrin receptor in proliferating and quiescent macrophages. Cell. L. E., and Wagner, O. (2012) The sedoheptulose kinase CARKL directs Immunol. 130, 401–415 macrophage polarization through control of glucose metabolism. Cell 33. Rolfe, D. F., and Brown, G. C. (1997) Cellular energy utilization and molec- Metab. 15, 813–826 ular origin of standard metabolic rate in mammals. Physiol. Rev. 77, 49. Vats, D., Mukundan, L., Odegaard, J. I., Zhang, L., Smith, K. L., Morel, 731–758 C. R., Wagner, R. A., Greaves, D. R., Murray, P. J., and Chawla, A. (2006) 34. Mazurek, S., and Eigenbrodt, E. (2003) The tumor metabolome. Anticancer Oxidative metabolism and PGC-1beta attenuate macrophage-mediated Res. 23, 1149–1154 inflammation. Cell Metab. 4, 13–24 35. Becker, L., Liu, N. C., Averill, M. M., Yuan, W., Pamir, N., Peng, Y., Irwin, 50. Yoshida, S., Tsutsumi, S., Muhlebach, G., Sourbier, C., Lee, M. J., Lee, S., A. D., Fu, X., Bornfeldt, K. E., and Heinecke, J. W. (2012) Unique pro- Vartholomaiou, E., Tatokoro, M., Beebe, K., Miyajima, N., Mohney, R. P., teomic signatures distinguish macrophages and dendritic cells. PloS Chen, Y., Hasumi, H., Xu, W., Fukushima, H., Nakamura, K., Koga, F., One 7, e33297 Kihara, K., Trepel, J., Picard, D., and Neckers, L. (2013) Molecular 36. Cook, A. D., Pobjoy, J., Steidl, S., Du¨ rr, M., Braine, E. L., Turner, A. L., chaperone TRAP1 regulates a metabolic switch between mitochondrial Lacey, D. C., and Hamilton, J. A. (2012) Granulocyte-macrophage colo- respiration and aerobic glycolysis. Proc. Natl. Acad. Sci. U.S.A. 110, ny-stimulating factor is a key mediator in experimental osteoarthritis pain E1604–1612 and disease development. Arthritis Res. Ther. 14, R199 51. Freemerman, A. J., Johnson, A. R., Sacks, G. N., Milner, J. J., Kirk, E. L., 37. O’Neill, L. A. (2014) Glycolytic reprogramming by TLRs in dendritic cells. Troester, M. A., Macintyre, A. N., Goraksha-Hicks, P., Rathmell, J. C., Nat. Immunol. 15, 314–315 and Makowski, L. (2014) Metabolic reprogramming of macrophages: 38. Everts, B., and Pearce, E. J. (2014) Metabolic control of dendritic cell glucose transporter 1 (GLUT1)-mediated glucose metabolism drives a activation and function: recent advances and clinical implications. Front. Immunol. 5, 203 proinflammatory phenotype. J. Biol. Chem. 289, 7884–7896 2732 Molecular & Cellular Proteomics 14.10

Journal

Molecular & Cellular Proteomics – American Society for Biochemistry and Molecular Biology

Published: Oct 1, 2015

There are no references for this article.