Integrating Phosphoproteome and Transcriptome Reveals New Determinants of Macrophage Multinucleation
2014; Elsevier BV; Volume: 14; Issue: 3 Linguagem: Inglês
10.1074/mcp.m114.043836
ISSN1535-9484
AutoresMaxime Rotival, Jeong‐Hun Ko, Prashant K. Srivastava, Audrey Kerloc’h, Alex Montoya, Claudio Mauro, Peter Faull, Pedro R. Cutillas, Enrico Petretto, Jacques Behmoaras,
Tópico(s)Epigenetics and DNA Methylation
ResumoMacrophage multinucleation (MM) is essential for various biological processes such as osteoclast-mediated bone resorption and multinucleated giant cell-associated inflammatory reactions. Here we study the molecular pathways underlying multinucleation in the rat through an integrative approach combining MS-based quantitative phosphoproteomics (LC-MS/MS) and transcriptome (high-throughput RNA-sequencing) to identify new regulators of MM. We show that a strong metabolic shift toward HIF1-mediated glycolysis occurs at transcriptomic level during MM, together with modifications in phosphorylation of over 50 proteins including several ARF GTPase activators and polyphosphate inositol phosphatases. We use shortest-path analysis to link differential phosphorylation with the transcriptomic reprogramming of macrophages and identify LRRFIP1, SMARCA4, and DNMT1 as novel regulators of MM. We experimentally validate these predictions by showing that knock-down of these latter reduce macrophage multinucleation. These results provide a new framework for the combined analysis of transcriptional and post-translational changes during macrophage multinucleation, prioritizing essential genes, and revealing the sequential events leading to the multinucleation of macrophages. Macrophage multinucleation (MM) is essential for various biological processes such as osteoclast-mediated bone resorption and multinucleated giant cell-associated inflammatory reactions. Here we study the molecular pathways underlying multinucleation in the rat through an integrative approach combining MS-based quantitative phosphoproteomics (LC-MS/MS) and transcriptome (high-throughput RNA-sequencing) to identify new regulators of MM. We show that a strong metabolic shift toward HIF1-mediated glycolysis occurs at transcriptomic level during MM, together with modifications in phosphorylation of over 50 proteins including several ARF GTPase activators and polyphosphate inositol phosphatases. We use shortest-path analysis to link differential phosphorylation with the transcriptomic reprogramming of macrophages and identify LRRFIP1, SMARCA4, and DNMT1 as novel regulators of MM. We experimentally validate these predictions by showing that knock-down of these latter reduce macrophage multinucleation. These results provide a new framework for the combined analysis of transcriptional and post-translational changes during macrophage multinucleation, prioritizing essential genes, and revealing the sequential events leading to the multinucleation of macrophages. Macrophage multinucleation (MM) 1The abbreviations used are:MMmacrophage multinucleationMGCsmultinucleated giant cellsNTNnephrotoxic nephritisBMDMbone marrow derived macrophagesFDRfalse discovery rateTFtranscription factor2-DG2 Deoxy glucosePIPhosphatidyl-inositol. 1The abbreviations used are:MMmacrophage multinucleationMGCsmultinucleated giant cellsNTNnephrotoxic nephritisBMDMbone marrow derived macrophagesFDRfalse discovery rateTFtranscription factor2-DG2 Deoxy glucosePIPhosphatidyl-inositol. occurs when macrophages fuse to form a single cell containing multiple nuclei. This phenomenon plays an essential role in several homeostatic and pathological processes. In healthy subjects, macrophage multinucleation is required for the formation of osteoclasts able to resorb bone excesses and maintain bone homeostasis (1Teitelbaum S.L. Bone resorption by osteoclasts.Science. 2000; 289: 1504-1508Crossref PubMed Scopus (3075) Google Scholar, 2Tonna E.A. Origin of osteoclats from the fusion of phagocytes.Nature. 1963; 200: 226-227Crossref PubMed Scopus (15) Google Scholar). Macrophage multinucleation also plays a major role in inflammatory conditions such as sarcoidosis or tuberculosis through the formation of multinucleated giant cells (MGCs), effector cells of granulomatous reaction (3Brodbeck W.G. Anderson J.M. Giant cell formation and function.Curr. Opin. Hematol. 2009; 16: 53-57Crossref PubMed Scopus (166) Google Scholar, 4Van Maarsseveen T.C.M.T. Vos W. van Diest P.J. Giant cell formation in sarcoidosis: cell fusion or proliferation with nondivision?.Clin. Exp. Immunol. 2009; 155: 476-486Crossref PubMed Scopus (30) Google Scholar). macrophage multinucleation multinucleated giant cells nephrotoxic nephritis bone marrow derived macrophages false discovery rate transcription factor 2 Deoxy glucose Phosphatidyl-inositol. macrophage multinucleation multinucleated giant cells nephrotoxic nephritis bone marrow derived macrophages false discovery rate transcription factor 2 Deoxy glucose Phosphatidyl-inositol. Despite the central role of MM in various biological mechanisms, the basic knowledge of its molecular determinants and the pathways regulating multinucleation remain largely unexplored. Moreover the presence of common regulators of both MGC and osteoclast formation suggests the existence of common mechanisms for macrophage multinucleation across different cell types (5Helming L. Gordon S. Molecular mediators of macrophage fusion.Trends Cell Biol. 2009; 19: 514-522Abstract Full Text Full Text PDF PubMed Scopus (250) Google Scholar, 6Peng Q. Malhotra S. Torchia J.A. Kerr W.G. Coggeshall K.M. Humphrey M.B. TREM2- and DAP12-dependent activation of PI3K requires DAP10 and is inhibited by SHIP1.Sci. Signal. 2010; 3: ra38Crossref PubMed Scopus (209) Google Scholar, 7Vignery A. Macrophage fusion the making of osteoclasts and giant cells.J. Exp. Med. 2005; 202: 337-340Crossref PubMed Scopus (167) Google Scholar, 8Zhang C. Dou C. Xu J. Dong S. DC-STAMP, the key fusion-mediating molecule in osteoclastogenesis.J. Cell Physiol. 2014; 229: 1330-1335Crossref PubMed Scopus (56) Google Scholar). Cell multinucleation was previously investigated at a transcriptional level in osteoclasts (9Rho J. Altmann C.R. Socci N.D. Merkov L. Kim N. So H. Lee O. Takami M. Brivanlou A.H. Choi Y. Gene expression profiling of osteoclast differentiation by combined suppression subtractive hybridization (SSH) and cDNA microarray analysis.DNA Cell Biol. 2002; 21: 541-549Crossref PubMed Scopus (61) Google Scholar) and fusing rat alveolar macrophages (10Cui W. Ke J.Z. Zhang Q. Ke H.-Z. Chalouni C. Vignery A. The intracellular domain of CD44 promotes the fusion of macrophages.Blood. 2006; 107: 796-805Crossref PubMed Scopus (91) Google Scholar). This led to the identification of novel determinants of osteoclast multinucleation (11Lee S.-H. Rho J. Jeong D. Sul J.-Y. Kim T. Kim N. Kang J.-S. Miyamoto T. Suda T. Lee S.-K. Pignolo R.J. Koczon-Jaremko B. Lorenzo J. Choi Y. v-ATPase V0 subunit d2-deficient mice exhibit impaired osteoclast fusion and increased bone formation.Nat. Med. 2006; 12: 1403-1409Crossref PubMed Scopus (449) Google Scholar) and macrophage fusion (10Cui W. Ke J.Z. Zhang Q. Ke H.-Z. Chalouni C. Vignery A. The intracellular domain of CD44 promotes the fusion of macrophages.Blood. 2006; 107: 796-805Crossref PubMed Scopus (91) Google Scholar). The Wistar-Kyoto (WKY) strain has been extensively used for its unique susceptibility to nephrotoxic nephritis (NTN), a highly reproducible and macrophage-dependent model of experimentally induced crescentic glomerulonephritis that has a large genetic component (12Aitman T.J. Dong R. Vyse T.J. Norsworthy P.J. Johnson M.D. Smith J. Mangion J. Roberton-Lowe C. Marshall A.J. Petretto E. Hodges M.D. Bhangal G. Patel S.G. Sheehan-Rooney K. Duda M. Cook P.R. Evans D.J. Domin J. Flint J. Boyle J.J. Pusey C.D. Cook H.T. Copy number polymorphism in Fcgr3 predisposes to glomerulonephritis in rats and humans.Nature. 2006; 439: 851-855Crossref PubMed Scopus (570) Google Scholar, 13Behmoaras J. Bhangal G. Smith J. McDonald K. Mutch B. Lai P.C. Domin J. Game L. Salama A. Foxwell B.M. Pusey C.D. Cook H.T. Aitman T.J. Jund is a determinant of macrophage activation and is associated with glomerulonephritis susceptibility.Nat. Genet. 2008; 40: 553-559Crossref PubMed Scopus (74) Google Scholar, 14Behmoaras J. Smith J. D'Souza Z. Bhangal G. Chawanasuntoropoj R. Tam F.W.K. Pusey C.D. Aitman T.J. Cook H.T. Genetic loci modulate macrophage activity and glomerular damage in experimental glomerulonephritis.J. Am. Soc. Nephrol. JASN. 2010; 21: 1136-1144Crossref PubMed Scopus (21) Google Scholar). The NTN in the WKY rat is also characterized by the formation of glomerular MGCs but the specific role of these cells in the progression of the disease remains to be determined. We observed that when the bone marrow derived macrophages (BMDMs) from the NTN-susceptible WKY rats are cultured in vitro under normal conditions, they spontaneously form MGCs after 3 days as opposed to NTN-resistant Lewis (LEW) BMDMs, which show very little fusion (Fig. 1A). We used this in vitro model of spontaneous multinucleation of BMDMs to identify a new major determinant of spontaneous macrophage multinucleation in a back-cross population derived from WKY and LEW rats (15Kang H. Kerloc'h A. Rotival M. Xu X. Zhang Q. D'Souza Z. Kim M. Scholz J.C. Ko J.-H. Srivastava P.K. Genzen J.R. Cui W. Aitman T.J. Game L. Melvin J.E. Hanidu A. Dimock J. Zheng J. Souza D. Behera A.K. Nabozny G. Cook H.T. Bassett J.H.D. Williams G.R. Li J. Vignery A. Petretto E. Behmoaras J. Kcnn4 is a regulator of macrophage multinucleation in bone homeostasis and inflammatory disease.Cell Rep. 2014; 8: 1210-1224Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar) as well as recapitulating previously known genes (DCSTAMP (8Zhang C. Dou C. Xu J. Dong S. DC-STAMP, the key fusion-mediating molecule in osteoclastogenesis.J. Cell Physiol. 2014; 229: 1330-1335Crossref PubMed Scopus (56) Google Scholar), MMP9 (16MacLauchlan S. Skokos E.A. Meznarich N. Zhu D.H. Raoof S. Shipley J.M. Senior R.M. Bornstein P. Kyriakides T.R. Macrophage fusion, giant cell formation, and the foreign body response require matrix metalloproteinase 9.J. Leukoc. Biol. 2009; 85: 617-626Crossref PubMed Scopus (114) Google Scholar)) as part of a macrophage gene regulatory network. Here, we developed an integrative approach to identify and characterize pathways involved in macrophage multinucleation by taking advantage of strain dependent spontaneous formation of MGCs during macrophage differentiation. We combined genome-wide and label-free phosphoproteome by LC-MS/MS and transcriptome by RNA-sequencing (RNA-seq) to pinpoint the cellular pathways as well as key regulators of transcriptomic changes occurring during macrophage multinucleation. We provide a first account of the interplay between post-translational and transcriptomic modifications occurring during macrophage multinucleation and MGC formation. Total RNA was extracted from WKY and LEW bone-marrow derived macrophages at the indicated time points during differentiation (day 3 and day 5) using Trizol (Invitrogen) according to manufacturer's instructions with an additional purification step by on-column DNase treatment using the RNase-free DNase Kit (Qiagen, UK) to ensure elimination of any genomic DNA. The integrity and quantity of total RNA was determined using a NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific) and Agilent 2100 Bioanalyzer (Agilent Technologies, UK). One microgram of total RNA was used to generate RNA-seq libraries using TruSeq RNA sample preparation kit (Illumina, UK) according to the manufacturer's instructions. Briefly, RNA was purified and fragmented using poly-T oligo-attached magnetic beads using two rounds of purification followed by the first and second cDNA strand synthesis. Next, cDNA 3′ ends were adenylated and adapters ligated followed by 10 cycles of library amplification. Finally, the libraries were size selected using AMPue XP Beads (Beckman Coulter, UK) purified and their quality was checked using Agilent 2100 Bioanalyzer. Samples were randomized to avoid batch effects and libraries were run on a single lane per sample of the HiSeq 2000 platform (Illumina) to generate 100 bp paired-end reads. An average of 72 m reads coverage per sample was achieved (minimum 38 m). RNA-seq reads were aligned to the rat (rn4) reference genome using tophat 2. The average number of mapped was 67 m (minimum 36 m) corresponding to an average mapping percentage of 93%. Sequencing and mapping were quality controlled using standard tools provided in the fastQC software. Gene level read counts were computed using HT-Seq-count with "union" mode (17Anders S. Pyl P.T. Huber W. HTSeq–A Python framework to work with high-throughput sequencing data.Bioinformatics. 2015; 15: 166-169Crossref Scopus (11029) Google Scholar) and genes with less than 10 aligned reads across all samples were discarded prior to analysis leading to 15,155 genes. Clustering of samples corresponding to different conditions was done using Ward's methods based on Euclidian distance of scaled sample gene expression profile. Differential expression analyses between groups were conducted using edgeR (18Robinson M.D. McCarthy D.J. Smyth G.K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data.Bioinformatics. 2010; 26: 139-140Crossref PubMed Scopus (21040) Google Scholar) within each strain and time point using a 5% FDR for each comparison. In order to identify significant differences in gene expression variation across time-points between the WKY and LEW BMDMs, a generalized linear model was fitted on all samples including strain, time point, and the [(strain) x (time point)] interaction term. The potential interaction was then tested using a Likelihood Ratio test (comparing to a model without the interaction term). The giant cell specific signature was defined as the set of genes that fulfilled the following criteria: Significant interaction between strain and time point at 5% FDR; Over-expression at day 5 compared with day 3 in WKY at 5% FDR; No differential expression at 5% nominal p value in LEW between day 3 and day 5; Over-expression at day 5 in WKY compared with LEW at 5% FDR. Rat bone marrow derived macrophages (BMDMs) were cultured as previously described (14Behmoaras J. Smith J. D'Souza Z. Bhangal G. Chawanasuntoropoj R. Tam F.W.K. Pusey C.D. Aitman T.J. Cook H.T. Genetic loci modulate macrophage activity and glomerular damage in experimental glomerulonephritis.J. Am. Soc. Nephrol. JASN. 2010; 21: 1136-1144Crossref PubMed Scopus (21) Google Scholar). BMDMs were flushed from femur and tibia bones from rats and cultured in presence of L929-conditioned media for the indicated times in Lab-Tek chambers (Fisher Scientific, UK). To assess spontaneous MGC formation, macrophages were fixed at the indicated times using Reastain Quick Diff and MGC quantification was performed by counting the number of nuclei in 100 macrophages using light microscopy. To evaluate the effect of siRNA knockdown on macrophage multinucleation, WKY BMDMs were transfected for 48 h with siGENOME SMARTpool for either rat Lrrfip1 or Smarca4 or Dnmt1 (100 nm, GE Healthcare, UK). All transfections were performed in parallel with siGENOME nontargeting siRNA pool as the scrambled control siRNA using Dharmafect 1 (1:50, Dharmacon) as a transfection reagent in OPTIMEM medium (Invitrogen). The siRNA sequences used in the siGENOME SMARTpool for all transcripts are available upon request. All qRT-PCRs were performed with a Viaa 7 Real-Time PCR system (Life Technologies, UK). A two-step protocol was used beginning with cDNA synthesis with iScript select (Bio-Rad, UK) followed by PCR using SYBR Green Jumpstart Taq Ready Mix (Sigma). A total of 10 ng of cDNA per sample was used. All samples were amplified using a set of at least four biological replicates with three technical replicates used per sample in the PCR. Viia 7 RUO Software was used for the determination of Ct values. Results were analyzed using the comparative Ct method and each sample was normalized to the reference gene Hprt, to account for any cDNA loading differences. The primer sequences are available upon request. Cells in the flasks were washed twice with cold PBS supplemented with phosphatase inhibitors (1 mm Na3VO4 and 1 mm NaF) and lyzed in urea lysis buffer (8 m Urea in 20 mm HEPES, pH 8.0) supplemented with phosphatase inhibitors (1 mm Na3VO4; 1 mm NaF; 1 mm β-glycerol-phosphate; 2.5 mm Na4P2O7; and 1 μm okadaic acid). Cell extracts were then scraped from the flasks, transfer to 2 ml low protein binding eppendorf tubes and further homogenized by sonication (three pulses of 15 s). Insoluble material was removed by centrifugation at 20,000 × g for 5 min at 4 °C and proteins in the supernatants were quantified by Bradford. A total amount of 500 μg and 200 μg of protein/sample were reduced and alkylated by sequential incubation with 10 mm DTT and 10 mm Iodoacetamide for 15 min at room temperature for phosphopeptides and whole peptides, respectively. 20 mm HEPES, pH 8.0, was used to reduce urea concentration to 2 m and proteins were digested overnight at 37 °C with Immobilized tosyl-lysine chloromethyl ketone (TLCK)–trypsin [20 p-toluenesulfonyl-l-arginine methyl ester (TAME) units/mg]. Digestion was stopped by adding trifluoroacetic acid (TFA) to 1% final concentration and trypsin beads were removed by centrifugation. Peptide mixtures were then desalted by reverse phase chromatographic cartridges (Oasis HLB, Waters, Milford, MA), washed, and dried. For quantitative phosphoproteomics, phosphopeptides were enriched following a procedure previously described, with some modifications (19Montoya A. Beltran L. Casado P. Rodríguez-Prados J.-C. Cutillas P.R. Characterization of a TiO2 enrichment method for label-free quantitative phosphoproteomics.Methods. 2011; 54: 370-378Crossref PubMed Scopus (94) Google Scholar). Briefly, sample volumes after elution from Oasis cartridges were adjusted to 1 ml with glycolic acid solution (1 m Glycolic acid; 80% ACN; and 5% TFA) and 50 μl of TiO2 beads (50% slurry in 1% TFA) were added to the peptide mixture. After 5 min incubation at RT with rotation, TiO2 slurry was packed into preconditioned empty spin tips by centrifugation. Spin tips were conditioned with 200 μl of 100% ACN. TiO2 beads were sequentially washed with 200 μl of glycolic acid solution, 100 mm Ammonium acetate in 25% ACN, and 1% ACN. Phosphopeptides were then eluted four times with 50 μl of 5% NH4OH in 1% ACN. Following the removal of the insoluble material by centrifugation from the eluents, supernatants were snap frozen in dry ice for 15 min. Samples were dried in a speed-vac and stored at −80 °C for mass spectrometry. For quantitative proteomics, mass spectrometry was applied without TiO2 enrichment only on WKY and LEW BMDM whole peptide extracts (analyzed in technical duplicates) at the peak of multinucleation. For quantitative phosphoproteomics, each of the four samples were provided in biological triplicate (in single technical replicate) for mass spectrometry analysis. Phosphopeptide and whole peptide dried extracts were resuspended in 14 μl of reconstitution buffer (0.1% TFA containing 20 nm of an enolase digest) and 4 μl (phosphopeptides) and 5 μl (peptides) were loaded in a LC-MS/MS system. This consists of a nanoflow liquid chromatography (nanoLC, Ultimate 3000, Thermo Scientific) system coupled online to an LTQ-Orbitrap Velos mass spectrometer (Thermo Scientific). The nanoLC system delivered a flow of 8 μl/min (loading) onto a trap column (Thermo Scientific Acclaim Pepmap 100 with dimensions 100 μm internal diameter and 2 cm length, C18 reverse phase material with 5 μm diameter beads and 100Å pore size) in 98% water, 2% acetonitrile, and 0.1% TFA. Peptides were then eluted on-line to an analytical column (Thermo Scientific Acclaim Pepmap RSLC with dimensions 75 μm internal diameter and 25 cm length, C18 reverse phase material with 2 μm diameter beads and 100Å pore size) and separated using a gradient with conditions: initial 5 min with 4% B (96% A), then 120 min gradient 4–35% B, then 10 min isocratic at 100% B, then 5 min isocratic at 4% B (solvent A: 98% water, 2% acetonitrile, and 0.1% formic acid; solvent B: 20% water, 80% acetonitrile, and 0.1% formic acid). For phosphopeptide quantification, a LTQ-Orbitrap Velos mass spectrometer acquired full scan survey spectra (m/z 350–1500) with a 15,000 resolution at m/z 400. A maximum of the seven (phosphopeptides) most abundant multiply-charged ions registered in each survey spectrum were selected in a data-dependent manner, fragmented by collision induced dissociation (multi-stage activation enabled) with a normalized collision energy of 35% and scan in the LTQ (m/z 50–2000). This produced a duty cycle of 2.4 s. A dynamic exclusion was enabled with the exclusion list restricted to 500 entries, exclusion duration of 60 s and mass window of 10 ppm. Chromatographic peaks were about 30 s at the base, which ensured at least 10 data points per extracted ion chromatogram (XIC). For whole protein quantification, a Q Exactive mass spectrometer acquired full scan survey spectra (m/z 400–2000) with 70,000 resolution at m/z 400 was used. A maximum of the 12 most abundant multiply-charged ions registered in each survey spectrum were selected in a data-dependent manner, fragmented by higher-energy collision induced dissociation (HCD) with a normalized collision energy of 28%. This produced a duty cycle of ∼1.5 s. MS/MS resolution was 17,500. For phosphopeptide identification, Mascot Distiller v2.4.3.1 was used to smoothen and centroid the MS/MS data and Mascot v2.4.1 search engine was used to match peaks to peptides in proteins present in the SwissProt Database (SwissProt_2013Jan.fasta) restricted to rattus norvegicus entries (7,853 sequences) (20Perkins D.N. Pappin D.J. Creasy D.M. Cottrell J.S. Probability-based protein identification by searching sequence databases using mass spectrometry data.Electrophoresis. 1999; 20: 3551-3567Crossref PubMed Scopus (6771) Google Scholar). The process was automated with Mascot Daemon v2.4, mass tolerance was set to 10 ppm and 600 millimass units for precursor and fragment ions, respectively. Phosphorylation on Ser, Thr, and Tyr; PyroGlu on N-terminal Glu; and oxidation of Met were allowed in the search as variable modifications and carbamidomethyl Cys as fixed modification. Trypsin (cleaves C-terminal to Arg and Lys residues provided there is not a Pro C-term to the Arg/Lys residue) was selected as the digestion enzyme and two missed cleavages were allowed. Sites of modification are reported when they had delta scores >10. Delta scores were calculated as previously described (21Savitski M.M. Lemeer S. Boesche M. Lang M. Mathieson T. Bantscheff M. Kuster B. Confident phosphorylation site localization using the Mascot Delta Score.Mol. Cell. Proteomics. 2011; 10M110.003830 Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar). Otherwise the site of modification was deemed to be ambiguous; in such cases phosphopeptides are reported as the start-end residues within the protein sequence. Results from Mascot searches were deposited into the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository (22Vizcaíno J.A. Côté R.G. Csordas A. Dianes J.A. Fabregat A. Foster J.M. Griss J. Alpi E. Birim M. Contell J. O'Kelly G. Schoenegger A. Ovelleiro D. Pérez-Riverol Y. Reisinger F. Ríos D. Wang R. Hermjakob H. The PRoteomics IDEntifications (PRIDE) database and associated tools: status in 2013.Nucleic Acids Res. 2013; 41: D1063-D1069Crossref PubMed Scopus (1596) Google Scholar) with the dataset identifier Project accession PXD001269 and Project DOI: 10.6019/PXD001269. Phosphopeptides showing a Mascot expectancy of 0 via likelihood maximization. The use of a parametric fit to estimate the probability of missing values rather than the nonparametric estimate proposed by Karpievitch et al. is justified here by the fact the nonparametric estimate led to unrealistic values (πl < 0) because of the unconstrained interpolation that was performed. Our estimate on the other hand, enforces strictly positive values of πl thus allowing more reliable estimates of the missing-at-random probability πl. p values resulting from the quantitative models were adjusted for multiple testing using Benjamini-Hochberg correction and a 5% FDR threshold as used for significance. For each biological replicate, values of protein abundance exported from Mascot were averaged between technical replicates to yield the final measures of protein abundance. Protein abundances were then log transformed (adding an offset of 1 to avoid infinite values) and samples intensities were scaled to have identical total intensity. For each protein, differential protein abundance between WKY and LEW was assessed using a t test. Giant-cell specific phosphorylation signature was defined based on peptides that were detected in at least two samples in all conditions. In order to improve power of the enrichment analysis (see below) we used nonconservative statistical thresholds to identify giant-cell specific phosphorylation signature. Although this approach might yield increased false positives, this will not bias the subsequent enrichment analysis because false positives will by definition occur at random. In detail, phosphopeptides were considered as exhibiting giant-cell specific signature if they fulfilled any of the two following s
Referência(s)