Portal de Periódicos da CAPES

Phosphotyrosine Proteomic Study of Interferon α Signaling Pathway Using a Combination of Immunoprecipitation and Immobilized Metal Affinity Chromatography

2005; Elsevier BV; Volume: 4; Issue: 6 Linguagem: Inglês

10.1074/mcp.m400077-mcp200

1535-9484

Haiyan Zheng, Ping Hu, Douglas Quinn, Yan Wang,

interferon and immune responses

Tyrosine phosphorylation is a type of post-translational modification that plays a crucial role in signal transduction. Thus, the study of this modification at the proteomic level has great biological significance. However, because of the low abundance of tyrosine-phosphorylated proteins in total cell lysate, it is difficult to evaluate the dynamics of tyrosine phosphorylation at a global level. In this work, proteins carrying phosphotyrosine (pTyr) were first purified from whole cell lysate by immunoprecipitation using anti-pTyr monoclonal antibodies. After tryptic digestion, phosphopeptides were further enriched by IMAC and analyzed by LC-MS. Quantitative changes of tyrosine phosphorylation at the global level were evaluated using isotopic labeling (introduced at the methyl esterification step prior to IMAC). Using this double enrichment approach, we characterized interferon α (IFNα)-induced pTyr proteomic changes in Jurkat cells. We observed induced phosphorylation on several well documented as well as novel tyrosine phosphorylation sites on proteins involved in IFNα signal transduction, such as Tyk2, JAK1, and IFNAR subunits. A specific site on α-tubulin (Tyr-271) was observed to be phosphorylated upon treatment as well. Furthermore, our results suggest that LOC257106, a CDC42 GAP-like protein, is potentially involved in this pathway. Tyrosine phosphorylation is a type of post-translational modification that plays a crucial role in signal transduction. Thus, the study of this modification at the proteomic level has great biological significance. However, because of the low abundance of tyrosine-phosphorylated proteins in total cell lysate, it is difficult to evaluate the dynamics of tyrosine phosphorylation at a global level. In this work, proteins carrying phosphotyrosine (pTyr) were first purified from whole cell lysate by immunoprecipitation using anti-pTyr monoclonal antibodies. After tryptic digestion, phosphopeptides were further enriched by IMAC and analyzed by LC-MS. Quantitative changes of tyrosine phosphorylation at the global level were evaluated using isotopic labeling (introduced at the methyl esterification step prior to IMAC). Using this double enrichment approach, we characterized interferon α (IFNα)-induced pTyr proteomic changes in Jurkat cells. We observed induced phosphorylation on several well documented as well as novel tyrosine phosphorylation sites on proteins involved in IFNα signal transduction, such as Tyk2, JAK1, and IFNAR subunits. A specific site on α-tubulin (Tyr-271) was observed to be phosphorylated upon treatment as well. Furthermore, our results suggest that LOC257106, a CDC42 GAP-like protein, is potentially involved in this pathway. Tyrosine phosphorylation plays a critical role in signal transduction that affects all aspects of cell life (1Hunter T. Cooper J.A. Protein-tyrosine kinases.Annu. Rev. Biochem. 1985; 54: 897-930Google Scholar). Changes in tyrosine phosphorylation have significant effects on cell growth, differentiation, and death. Deregulation of these events may lead to diseases such as cancer and metabolic disorders. Tyrosine phosphorylation has been extensively studied by the conventional, one protein at a time approach. However, to fully understand the complex signaling dynamics, global study of this modification is important. It is estimated that only 0.05–0.1% of protein phosphorylation occurs on tyrosine residues (2Cooper J.A. Sefton B.M. Hunter T. Detection and quantification of phosphotyrosine in proteins.Methods Enzymol. 1983; 99: 387-402Google Scholar). Thus, it has been challenging to extensively document tyrosine phosphorylation events in a signaling pathway. Monoclonal antibodies that specifically recognize phosphotyrosine (pTyr) 1The abbreviations used are: pTyr, phosphotyrosine; GAP, GTPase-activating protein; IFN, interferon; IFNAR2c, interferon α/β receptor β chain; IFNAR1, interferon α/β receptor α chain; MWCO, molecular weight cut-off; RP, reverse phase; STAT, signal transducers and activators of transcription. 1The abbreviations used are: pTyr, phosphotyrosine; GAP, GTPase-activating protein; IFN, interferon; IFNAR2c, interferon α/β receptor β chain; IFNAR1, interferon α/β receptor α chain; MWCO, molecular weight cut-off; RP, reverse phase; STAT, signal transducers and activators of transcription. have been widely used in research related to protein tyrosine phosphorylation. Tyrosine-phosphorylated proteins can be effectively immunoprecipitated from cell lysate by these antibodies. Conventionally, after immunoprecipitation, proteins are separated and quantified by SDS-PAGE or two-dimensional PAGE. The protein spots bearing differences between the control and the treated samples are identified using in-gel proteolysis and MS. Phosphopeptides can be selectively enriched from in-gel digested samples by IMAC if phosphorylation sites are to be identified (3Schlosser A. Bodem J. Bossemeyer D. Grummt I. Lehmann W.D. Identification of protein phosphorylation sites by combination of elastase digestion, immobilized metal affinity chromatography, and quadruple-time of flight tandem mass spectrometry.Proteomics. 2002; 2: 911-918Google Scholar). Alternatively, parent ion scanning of phosphate- or pTyr-specific marker fragment ions may be used (4Steen H. Fernandez M. Ghaffari S. Randely A. Mann M. Tyrosine phosphorylation mapping of epidermal growth factor receptor signaling pathway.J. Biol. Chem. 2003; 277: 1031-1039Google Scholar) to detect phosphopeptides in the in-gel-digested samples, and phosphorylation sites can be identified by MS/MS. However, these gel-based approaches are not very effective (5Neubauer G. Mann M. Mapping of phosphorylation sites of gel-isolated proteins by nanoelectrospray tandem mass spectrometry: Potentials and limitations.Anal. Chem. 1999; 71: 235-242Google Scholar) and are not amenable to high throughput studies. In contrast, solution-based phospho-proteomics are preferred. Rapid advances have been made in the area of technology development regarding phosphopeptide enrichment in recent years. In particular, methyl esterification of acidic side chains of peptides has been demonstrated to dramatically reduce the nonspecific background in IMAC-based approaches. This strategy has allowed for the identification of over 200 phosphopeptides from a yeast lysate (6Ficarro S.B. McCleland M.L. Stukenberg P.T. Burke D.J. Ross M.M. Shabanowitz J. Hunt D.F. White F.M. Phosphoproteome analysis by mass spectrometry and its application to Saccharomyces cerevisiae.Nat. Biotechnol. 2002; 20: 301-305Google Scholar). However, due to the presence of the overwhelming serine/threonine phosphorylation, it is still difficult to assess the global dynamics of tyrosine phosphorylation using this method alone. Enrichment of pTyr-containing proteins prior to IMAC enrichment has been shown to significantly improve the detection of tyrosine phosphorylation (7Salomon A.R. Ficarro S.B. Brill L.M. Brinker A. Phung Q.T. Ericson C. Sauer K. Brock A. Horn D.M. Schultz P.G. Peters E.C. Profiling of tyrosine phosphorylation pathways in human cells using mass spectrometry.Proc. Natl. Acad. Sci. U S A. 2003; 100: 443-448Google Scholar). Interferons (IFNs) are a group of secreted proteins that belong to the cytokine family. They are classified as type I IFNs (including IFNα, IFNβ, and IFNω) and type II IFN (IFNγ). Type I interferons, such as IFNα and IFNβ, are produced by numerous types of cells, including leukocytes and fibroblasts, that elicit different effects depending on the type of cell that receives the signal (8Diaz M.O. Pomykala H.M. Bohlander S.K. Maltepe E. Malik K. Brownstein B. Olopade O.I. Structure of the human type-I interferon gene cluster determined from a YAC clone contig.Genomics. 1994; 22: 540-552Google Scholar). The anti-viral, anti-proliferative, and immuno-modulatory activities of IFNα have been utilized in treatment for chronic myelogenous leukemia and hepatitis B and C for many years (9Goldman J.M. Druker B.J. Chronic myeloid leukemia: Current treatment options.Blood. 2001; 98: 2039-2042Google Scholar, 10Carreno V. Castillo I. Molina J. Porres J.C. Bartolome J. Long-term follow-up of hepatitis B chronic carriers who responded to interferon therapy.J. Hepatol. 1992; 15: 102-106Google Scholar, 11Liang T.J. Rehermann B. Seeff L. Hoofnagle J. Pathogenesis, natural history, treatment, and prevention of hepatitis C.Ann. Intern. Med. 2000; 132: 296-305Google Scholar). It is known that IFNα elicits its effect by binding to specific membrane receptors (IFNAR1 and IFNAR2c) on the cell surface and activates the receptor-associated tyrosine kinases, which in turn activate a series of transcription factors known as signal transducers and activators of transcription or STAT (12Stark G.R. Kerr I.M. Williams B.R.G. Silverman R.H. Schreiber R.D. How cells respond to interferons.Annu. Rev. Biochem. 1998; 67: 227-264Google Scholar). However, the mechanistic details of IFNα function are still poorly understood. Evaluation at a global scale of tyrosine phosphorylation events involved in the signaling of IFNα will help us gain insight of this important pathway. This article describes an improved double enrichment method that allows quantitative study of signaling events in the aspect of tyrosine phosphorylation. Using this method, we studied IFNα signaling pathway in Jurkat cells. Tyrosine-phosphorylated proteins were immunoprecipitated from control and treated Jurkat cell lysates using a mixture of two pTyr-specific monoclonal antibodies. Furthermore, tryptic digest of the immunoprecipitated proteins was methyl-esterified, and phosphopeptides were enriched by IMAC and analyzed by reverse phase (RP) LC-MS/MS. Protein and phosphorylation site identification was achieved by data base search using the MS/MS data. For quantitation, stable isotope labeling was employed. In this case, after immunoprecipitation and tryptic digestion, peptides from treated and control samples were methyl-esterified by H3- or D3-methanolic HCl, respectively. The two isotope-labeled samples were then mixed prior to IMAC. Enriched phosphopeptides were separated by HPLC and detected by MS. Extracted ion chromatograms of the hydrogen and deuterium forms of the peptide ions were compared to obtain relative quantitation between control and treated samples (see Fig. 1). Using this approach, important regulatory factors in IFNα signaling events were identified, and the induced tyrosine phosphorylation by the treatment was observed and quantified. Also revealed by this study was a site of tyrosine phosphorylation of α-tubulin that was affected by IFNα treatment. Also, for the first time we showed that a putative CDC42 GTPase-associated protein (GAP)-like protein may be involved in the pathway. Our results demonstrated the feasibility of this method in studying signaling events by characterization of protein tyrosine phosphorylation globally. Jurkat clone E6–1 was obtained from American Type Culture Collection. Cells were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum, 2 mm l-glutamine, 100 μg/ml streptomycin, and 100 units/ml penicillin G (all from Invitrogen) in a 5.0% CO2 incubator at 37 °C. Subconfluent culture (1 × 106 cells/ml) was harvested, washed with fresh medium once, and then resuspended in fresh medium (prewarmed to 37 °C) at a density of 0.5 × 108 cells/ml. IFNα (Research Diagnostics Inc., Flanders, NJ) or an equal volume of fresh medium (control sample) was added to the cells and incubated at 37 °C for 5 min before lysis with the addition of 5× lysis buffer (1× lysis buffer: 50 mm Tris-HCl, pH 7.5, 1 mm EDTA, 150 mm NaCl, 1% Nonidet P-40, 2 mm Na3VO4, 2 mm NaF, and protease inhibitor mixture (Roche Diagnostics)). Lysate was first cleared by centrifugation at 8,000 × g for 10 min at 4 °C and filtered through a 0.45-μm syringe filter. 1:1 Monoclonal anti-pTyr agarose PT66 (Sigma-Aldrich) and 4G10 (Upstate Cell Signaling Solutions, Waltham, MA) were added to the lysate at a ratio of 200 μl of resin slurry/1 × 109 cells and mixed for 4–15 h at 4 °C. Beads were washed twice with 1× lysis buffer, and proteins were eluted twice with 400 μl of 8 m urea in 50 mm NH4HCO3 in the presence of 20 mm ethylamine for 5 min each at 96 °C, and the eluates were combined. Proteins were reduced by incubation in the presence of 10 mm DTT at 96 °C for 10 min. After being cooled to room temperature, sulfhydryl groups were alkylated by reacting with 20 mm iodoacetamide in the dark for 30 min. The buffer was changed to 50 mm NH4HCO3 via dialysis using a 10-kDa molecular weight cut-off (MWCO) Slide-A-Lyzer (Pierce). Trypsin (modified sequencing grade; Promega, Madison, WI) digestion was performed at ∼1:50 to 1:100 trypsin-to-protein ratio (w/w) overnight at 37 °C for 16 h. The tryptic digest was filtered through a spin filter of 10-kDa MWCO (Millipore, Bedford, MA) and speed-vac dried. H3-methanolic HCl or D3-methanolic HCl was prepared by adding 160 μl of acetyl chloride to 1 ml of methyl H3-alcohol or methyl D3-alcohol while stirring. Immunoprecipitated proteins from starting material of 1 × 109 cells were treated with 200 μl of H3/D3-methanolic HCl at room temperature for 30 min and then speed-vac dried. Poros MC column (2.1 × 30 mm, 100-μl bed volume; Applied Biosystems, Foster City, CA) was activated by the loading of 0.2 m FeCl3. After conditioning with wash buffer 1 (1% acetic acid) and wash buffer 2 (1% acetic acid in 50% ACN), methyl-esterified peptides were solubilized in wash buffer 2 and loaded onto the column. The column was extensively washed with wash buffer 2 prior to the elution of phosphopeptides using 2% NH4OH in 50% ACN (v/v). The eluate was neutralized to pH 5 with the addition of acetic acid and then speed-vac dried. For the global phosphorylation study of Ser, Thr, and Tyr phosphorylation, cells were treated and lysed as described above. The protein concentration was measured using Bio-Rad DC Protein Assay. Proteins were digested, and the methyl esterification/IMAC procedure was performed as described above. For peptide identification, control and treated samples were analyzed by LC-MS/MS. Quantitative comparison of phosphorylation was achieved through inverse labeling of control and treated samples (13Wang Y.K. Ma Z. Quinn D.F. Fu E.W. Inverse 18O labeling mass spectrometry for the rapid identification of marker/target proteins.Anal. Chem. 2001; 73: 3742-3750Google Scholar). Specifically, half of each sample was methyl esterified using H3-methanolic HCl and the other half using D3-methanolic HCl. After labeling, the hydrogen form of control sample was mixed with the deuterium form of treated sample, and the deuterium form of control sample was mixed with the hydrogen form of treated sample. After IMAC, the enriched phosphopeptides were analyzed by LC-MS. Peptides eluted off IMAC were separated by RP-HPLC using Pepmap C18 (3 μm, 100 Å, 180 μm × 15 cm; Dionex, Surrey, UK) with a gradient of 5–40% B in 60 min at a flow rate of 2 μl/min (0.1% formic acid in water as A and 0.1% formic acid in ACN as B). Peptides were directly eluted to a Q-TOF Ultima mass spectrometer (Waters, Manchester, UK) for on-line MS analysis. For peptide sequence and phosphorylation site determination, the data-dependent LC-MS/MS analysis was performed separately on control and treated samples using the following settings: MS survey scan from m/z 400–1,800 was followed by MS/MS on the three most intense multiply charged ions (two to four charges) that were above a preset intensity threshold with dynamic exclusion of 90 s. Collision energy was automatically determined by the charge state and the m/z value of a particular precursor ion. MASCOT software (Matrix Science, London, UK) was used to search the MS/MS data against the NCBI data base (human) for peptide sequence/parent protein identification and phosphorylation site determination. The mass tolerance of both the precursor ions and the MS/MS fragment ions was set at ± 0.2 Da. Whereas carbamidomethyl cysteine and methyl esterification of the C terminus and aspartic and glutamic acids were set as static modifications, oxidation of methionine and phosphorylation of serine, threonine, and tyrosine were set as variable modifications. MS/MS spectra were checked manually to verify the sequence assignments. In the quantitative analysis, the mixtures of isotope-labeled control and treated samples were analyzed under similar LC-MS conditions but without data-dependent MS/MS. The ratio of the integrated peak areas of an isotopic pair was used to obtain quantitative comparison of the phosphopeptide between treated and control samples. IFNα is known to bind to two membrane receptor subunits and activates the JAK/STAT signal transduction pathway by tyrosine phosphorylation of JAK1, Tyk2, STAT1, and STAT2 (12Stark G.R. Kerr I.M. Williams B.R.G. Silverman R.H. Schreiber R.D. How cells respond to interferons.Annu. Rev. Biochem. 1998; 67: 227-264Google Scholar). Published protocols vary in treatment conditions. Thus, pilot experiments were conducted to determine a treatment condition that was suitable for the study. We tested the treatment duration varying from 5 min to 1 h (see Fig. 2A) and also at various concentrations of IFNα (Fig. 2C). Using STAT1 (pTyr701) as a marker, we found that the treatment of Jurkat cells with IFNα at a concentration of 1 × 104 units/ml for 5–10 min led to sufficient induction of tyrosine phosphorylation. Thus, a 5-min treatment condition with 1 × 104 units/ml IFNα was chosen for the study. Next, we evaluated 4G10 and PT66, two strains of monoclonal antibodies against pTyr that have been widely used for Western blot and immunoprecipitation. Proteins in Jurkat cell lysate (control and IFNα-treated) were separated on SDS-PAGE and transferred to nitrocellulose membranes. Western blot using 4G10 or PT66 was performed following the manufacturer's recommended conditions (Fig. 2B). The different protein profiles on Western blots using 4G10 and PT66 indicated that these two antibodies had differences in affinity/recognition toward tyrosine-phosphorylated proteins in the lysate. Thus, we decided to use the mixture of the two antibodies (at 1:1 ratio) for immunoprecipitation to achieve better coverage for pTyr. Total tyrosine phosphorylation was evaluated by Western blot using the same combination of anti-pTyr antibodies that were used for immunoprecipitation. It was difficult to detect any difference between control and treated samples (Fig. 2A). This result suggested that the induced changes in tyrosine phosphorylation occurred on a small number of specific proteins rather than massively on a large number of proteins such as those induced by sodium pervanadate treatment (a nonspecific tyrosine phosphatase inhibitor, data not shown). After anti-pTyr immunoprecipitation, tryptic digestion, methyl esterification, and IMAC enrichment, the isolated peptides were analyzed by capillary HPLC on-line with MS using a Q-TOF Ultima instrument as described in "Material and Methods." The MS/MS spectra were used to search the NCBI human data base using MASCOT to identify the amino acid sequence and phosphorylation site(s) (Fig. 3). Each search result of a MS/MS spectrum was manually confirmed. Using this method, we identified 38 tyrosine phosphorylation sites and 19 phosphoserine/threonine sites (Table I). Some of the phosphoserine/threonine-containing peptides were recovered because the protein was tyrosine phosphorylated. For example, three peptides were recovered for LAT protein. While one was tyrosine phosphorylated, two were phosphorylated at Ser sites (EYVNpSQELHPGAAK (Ser-84) and pSPQPLGGSHR (Ser-195)). These two peptides were detected because LAT was tyrosine phosphorylated (EpYVNVSQELHPGAAK (Tyr-191)) (Table I). We also detected several peptides of serine phosphorylation but failed to identify any tyrosine phosphorylation of the corresponding proteins. One possible reason was that the tyrosine phosphorylation site(s) might be in a sequence region that was difficult to map using trypsin digestion and LC-MS. Another possible scenario was that the protein formed a complex with a tyrosine-phosphorylated protein and was co-precipitated with the complex. It was also possible that they represented the nonspecific background in immunoprecipitation. Further investigation is required before a conclusion can be made. In addition, nonphosphorylated peptides from immunoglobin were also detected (Fig. 4D), which was likely the result of antibody leaching off the immuno-beads. In comparison, when general phosphorylation was evaluated without pre-enrichment of pTyr proteins, less than 1% of the peptides identified were pTyr-containing peptides (data not shown), and they corresponded to proteins of relatively high abundance. For example, using 107 cells, Tyr-15 phosphorylation of CDC2 (IGEGTpYGVVYK) was detected using IMAC phosphopeptide enrichment alone without the anti-pTyr immunoprecipitation (data not shown). However, when the double-enrichment method described here was applied, this peptide was not detected at a signal intensity proportional to the amount of starting material (109 cells). Recognizing that antibody capacity was not the limiting factor, the phenomenon suggested that anti-pTyr antibodies, in addition to the recognition of pTyr, had a degree of preference for certain groups/sequences of proteins. Thus, the detection of a tyrosine phosphorylation using this method may not be a true representation of the abundance of pTyr in the cell.Table IPhospho-peptide identification and relative quantitationProteinNCBl gi numberPeptide sequence (phosphorylation site on protein)Treated: controlBcl-2-associated transcription factorgi|28838345STFREEpSPLR (S358)0.98CD3 δ chaingi|7459632DDAQpYSHLGGNWAR (Y160)0.94CD3 δ chaingi|7459632DRDDAQpYSHLGGINWAR (Y160)0.92CD3 δ chaingi|7459632NDQVPpYQPLR (Y149)0.76CD3 εgi|91760DLpYSGLNQR (Y199)1.21CD3 εgi|91760NDQVpYQPLRDR (Y149)0.97CD3 εgi|91760ERPPPVPNPDpYEPIR (Y188)1.01CD3 εgi|91760ERPPPVPNPDpYEPIRK (Y188)0.90CD3-ζgi|4557431DTpYDALHMQALPPR (Y153)1.03CD3-ζgi|4557531EEpYDVLDKR (Y83)0.92CD3-ζgi|4557531REEpYDVLDKR (Y83)0.94CD3-ζgi|4557431GHDGLpYQGLSTATK (Y141)0.98CD3-ζgi|4557431GKGHDGLpYQGLSTATK (Y141)1.11CD3-ζgi|4557431KNPQEGLpYNELQK (Y110)1.10CD3-ζgi|4557431RKNPQEGLpYNELQK (Y110)1.09CD3-ζgi|4557431MAEApYSEIGMK (Y122)0.93CD3-ζgi|4557431SADAPAYQQGQNQLpYNELNLGR (Y72)0.89CD5gi|7656965SHAENPTASHVDNEpYSQPPR (Y453)0.78CDK2gi|29849IGEGTpYGVVYK (Y15)1.09DNA-binding proteingi|181914NEEDEGHSNpSSPR (S61)0.90Docking protein 1: p62dokgi|4503357SHNSALpYSQVQK (Y449)1.02Drebrin 1 isoform bgi|18426913LpSSPVLHR (S143)1.38Fyn/Lckgi|34289LIEDNEpYTAR (Y27)0.83Fyn/Lckgi|34289NLONGGFpYISPR (Y192)1.14IGF-II mRNA-binding protein 1gi|4191608QGpSPVAAGAPAK (S181)1.40Interferon α/β receptor α subunitgi|32672NLLLSTpSEEQIEK (S495)>5Interferon α/β receptor β subunitgi|32672RKpSPLQDPFPEEDpYSSTEGSGGR (S400, Y411)>5Interferon α/β receptor β subunitgi|32672RKpSPLQDPFPEEDYSSTEGSGGR (S400)>5Intersectin 2gi|22325381REEPEALpYAAVNK (Y24)1.13JAK1gi|107499pYIPETLNK (Y220)>5JAK1gi|107499AIETDKEpYpYTVKDDR (Y1034, Y1035)>5KIAA0699 proteingi|3327212pSFILLFK (S574)0.8LATgi|2780223EYVNVpSQELHPGAAK (S195)1.19LATgi|2780223pSPQPLGGSHR (S84)1.15LATgi|2780223EpYYNVSQELHPGAAK (Y191)1.09LATgi|2780223LPGSYDpSTSSDSLYPR (S39/T40/S41)1.31LIM lipomagi|5031887YYEGpYYAAGPGYGGR (Y301)1.22LOC257106gi|27497821GpSGSLEGEAAGCGR (S630)>5LOC257106gi|27497821GpSGpSLEGEAAGCGR (S630, S632)>5LPP-PRFT fusion proteingi|21668380YYEGpYYAAGPGYGGR (Y25)0.78p38 MAPKgi|2554639HTDDEMTGpYVATR (Y188)1.01Phosphoprotein associated with GEMsgi|7682684AEFAEpYApSVDR (Y227, S229)0.75Phosphoprotein associated with GEMsgi|7682864AEFAEpYASVDR (Y227)1.20Phosphoprotein associated with GEMsgi|7662864CRQpSVNVESILGNSCDPEEEAPPPVPVK (S254)0.90Phosphoprotein associated with GEMsgi|7682684ENDpYESISDLQQGR (Y417)0.73Phosphoprotein associated with GEMsgi|7682684SGQSLTVPESTpYTSIQGDPQR (Y341)1.06Phosphoprotein associated with GEMsgi|7682684SPSSCNDLpYATVK (Y359)1.10Phosphoprotein associated with GEMsgi|7682684SPSSCNDLpYATVKDFEK (Y359)1.16Ras GAP SH3 binding proteingi|5031703SSSPAPADIAQTVQEDLR (S230/S231/S232)1.21Rho GEF 7gi|288005MpSGFIYQGK (S165)0.99SH3P18gi|1438935REEPEALpYAAVNK (Y24)1.13SHP2 interacting transmembrane adaptorgi|7657577pYSEVVLDSEPK (Y148)0.77TRAP150gi|4530441ASAVSELpSPR (S243)0.75TRAP150gi|4530441IDIpSPSTFR (S682)0.88α-Tubulingi|32015IHFPLATpYAPVISAEK (Y271)>5Tyk2gi|31543838LLAQAEGEPCpYIR (Y292)>5Tyk2gi|31543838AVPEGHEpYpYRVR (Y1054, Y1055)>5ZAP70gi|346421ALGADDSpYpYTAR (Y492, Y493)1.09ZAP70gi|346421ALGADDSpYYTAR (Y492/Y493)1.11ZAP70gi|346421LKADGLIpYCLK (Y248)1.10ZAP70gi|346421IDTLNSDGpYTPEPAR (Y292)0.98 Open table in a new tab Fig. 4Extracted ion chromatograms of hydrogen:deuterium pairs and the relative quantitation. In this quantitative analysis, hydrogen form was originated from IFNα-treated sample and deuterium from control sample. A, RKpSPLQDPFPEEDYSSTEGSGGR (Ser-400) IFNAR2c; B, LLAQAEGEPCpYIR (Tyr-292) Tyk2; C, GpSGSLEGEAAGCGR (Ser-630) LOC257106; D, HNSYTCEATHK Fab (IgG).View Large Image Figure ViewerDownload (PPT) Several phosphopeptides were exclusively detected in the IFNα-treated samples (Table I). LLAQAEGEPCpYIR (Tyr-292) and AVPEGHEpYpYRVR (Tyr-1054, Tyr-1055) are peptides from Tyk2, AIETDKEpYpYTVKDDR (Tyr-1034, Tyr-1035) and pYIPETLNK (Tyr-220) belong to JAK1, whereas RKpSPLQDPFPEEDpYSSTEGSGGR (Ser-400, Tyr-411) and RKpSPLQDPFPEEDYSSTEGSGGR (Ser-400) are from IFNα/β receptor β chain (IFNAR2c), and NLLLSTpSEEQIEK (Ser-495) is a peptide from IFNα/β receptor α chain (IFNAR1). These four proteins are known to be involved in IFNα signal transduction (14Barbieri G. Velazquez L. Scrobogna M. Fellous M. Pellegrini S. Activation of the protein tyrosine kinase tyk2 by interferon α/β.Eur J Biochem. 1994; 223: 427-435Google Scholar). Tyr-411 of IFNAR2c is at the JAK2 binding domain (15Nadeau O.W. Domanski P. Usacheva A. Uddin S. Platanias L.C. Pitha P. Raz R. Levy D. Majchrzak B. Fish E. Colamonici O.R. The proximal tyrosines of the cytoplasmic domain of the β chain of the type I interferon receptor are essential for signal transducer and activator of transcription (Stat) 2 activation. Evidence that two Stat2 sites are required to reach a threshold of interferon α-induced Stat2 tyrosine phosphorylation that allows normal formation of interferon-stimulated gene factor 3.J. Biol. Chem. 1999; 274: 4045-4052Google Scholar). Tyr-1054 and Tyr-1055 of Tyk2 are important for signal transduction of IFNα (16Gauzzi M.C. Velazquez L. McKendry R. Mogensen K.E. Fellous M. Pellegrini S. Interferon-α-dependent activation of Tyk2 requires phosphorylation of positive regulatory tyrosines by another kinase.J. Biol. Chem. 1996; 271: 20494-20500Google Scholar). Tyr-292 of Tyk2, however, is a novel site that has not been documented previously. To quantitatively compare changes of tyrosine phosphorylation upon treatment of IFNα, we incorporated stable isotope labeling at the methyl esterification step. Treated sample was labeled as hydrogen form and control sample as deuterium form. The key in performing quantitation using isotope labeling was in paring up the isotopic peaks, which proved to be challenging in this case as variable mass differences were produced between the pair (depending on the peptide sequence). In addition, deuterium labeling led to deviation in elution time on RP LC. For peptides for which identification had been established, the isotopic peak correlation was achieved by the predictive calculation of the mass of the deuterium form based on the sequence (i.e. the number of acidic residues). Additionally, the LC elution characteristics (deuterium form elutes slightly earlier than the hydrogen form) (17Goodlett D.R. Keller A. Watts J.D. Newitt R. Yi E.C. Purvine S. Eng J.K. von Haller P. Aebersold R. Kolker E. Differential stable isotope labeling of peptides for quantitation and de novo sequence derivation.Rapid Comm. Mass Spec. 2001; 15: 1214-1221Google Scholar) were taken into consideration. Alternatively, without the knowledge of the peptide identity, the isotope peak pairing could be achieved by applying the plausible hydrogen and deuterium form mass differences, the C13 isotope patterns, and the LC elution characteristics. Once the isotopic peak pairs were correlated, the integrated peak areas were used to calculate the intensity ratio and to achieve the quantitative comparison of every phosphopeptide between the treatment and control samples (Table I). In this study, the ratio of hydrogen to deuterium form for most peptides fluctuated within the range of 0.75–1.25 or within 1.5-fold. However, for Tyk2, JAK1, IFNAR2c, and IFNAR1 peptides, hydrogen:deuterium ratios were greater than 5:1, representing a significant change upon treatment (Fig. 4, A and B). These results were consistent with the known involvement of the proteins in IFNα signal transduction. An α-tubulin peptide, IHFPLATpYAPVISAEK (Tyr-271), was also found to be present exclusively in the IFNα-treated sample. Furthermore, we identified a CDC42 GAP-like protein (LOC257106) with high confidence by two phosphoserine-containing peptides, GpSGSLEGEAAGCGR (Ser-630) and GpSGpSLEGEAAGCGR (Ser-630, Ser-632), which showed a hydrogen:deuterium ratio of >5 (Fig. 4C). Our data suggested that this protein was likely involved in the signal transduction of IFNα. Because neither of these two peptides contained a pTyr, we further investigated whether IFNα treatment affected the phosphorylation of the identified phosphoserine sites by a general phosphorylation analysis (IMAC enrichment alone). Only the monophosphorylated pS630 peptide was reliably detected in the general phosphorylation analysis (the doubly phosphorylated peptide was detected at lower intensity when both were detected in the pTyr proteomic analysis). No quantitative difference was observed for this peptide between IFNα-treated and control samples (Fig. 5), suggesting that at least the phosphorylation of Ser-630 was not affected by IFNα treatment. We examined the feasibility of using pTyr-specific antibodies to enrich tyrosine-phosphorylated proteins. In combination with proteolysis, methyl esterification, IMAC, and isotope labeling, we were able to detect tyrosine phosphorylation induced by IFNα treatment in Jurkat cells. Using 1 × 109 cells, we identified proteins that were known to be involved in signal transduction of IFNα. Among them, a previously undocumented tyrosine phosphorylation site, Tyr-292 of Tyk2, was identified to be involved in the signaling pathway. In addition, we have identified α-tubulin and a CDC42 GAP-like protein that have not been previously shown to be involved in this pathway. Two proteins (STAT1 and STAT2) with tyrosine phosphorylation sites that were previously reported to be influenced by IFNα treatment were not detected in this study. Whereas it is possible that restricted accessibility of antibody to the phosphorylation sites may have limited the recovery by immunoprecipitation, a similar STAT complex was recovered using an anti-pTyr immunoprecipitation procedure (18Improta T. Schindler C. Horvath C.M. Kerr I.M. Stark G.R. Darnell Jr., J.E. Transcription factor ISGF-3 formation requires phosphorylated Stat91 protein, but Stat113 protein is phosphorylated independently of Stat91 protein.Proc. Natl. Acad. Sci. U S A. 1994; 91: 4776-4780Google Scholar, 19Cruz-Vera J. Clara L. Hernandez-Kelly R. Mendez J.A. Perez-Salazar E. Ortega A. Collagen-induced STAT family members activation in Entamoeba histolytica trophozoites.FEMS Microbiol. Lett. 2003; 229: 203-209Google Scholar). More likely, it is possible that the abundance of the STAT1 and STAT2 proteins falls below the detection limits of our experiment. A recent publication in which a phosphotyrosine proteomic study was carried out on Jurkat cells treated with sodium pervanadate (a nonspecific tyrosine phosphatase inhibitor) supports this rationale (20Brill L.M. Salomon A.R. Ficarro S.B. Mukherji M. Stettler-Gill M. Peters E.C. Robust phosphoproteomic profiling of tyrosine phosphorylation sites from human T cells using immobilized metal affinity chromatography and tandem mass spectrometry.Anal. Chem. 2004; 76: 2763-2772Google Scholar). pTyr sites on STAT1/2 were not detected in the study. In the case of the STAT2 protein, the primary sequence of the tryptic peptide may prevent its detection. The pTyr-690 site has a surrounding sequence of NLQERRKpYLKHRLIV (21Gurney A.L. Wong S.C. Henzel W.J. De Sauvage F.J. Distinct regions of c-Mpl cytoplasmic domain are coupled to the JAK-STAT signal transduction pathway and Shc phosphorylation.Proc. Natl. Acad. Sci. U S A. 1995; 92: 5292-5296Google Scholar). Upon trypsin digestion the pTyr-690 site is located in a hydrophilic trimer peptide (pYLK), which would be difficult to detect under the RP LC/MS condition. Among identified peptides, less than 8% were nonphosphorylated, all of which were derived from immunoglobin (leached from the antibody column). In contrast, rarely any nonphosphorylated peptide was detected in the general phospho-proteomic study using IMAC alone. It was possible that in an eluate of immunoprecipitation, the leached antibody protein was present at a substantial proportion relative to other protein components and caused a background in the IMAC enrichment. Ficarro et al. (6Ficarro S.B. McCleland M.L. Stukenberg P.T. Burke D.J. Ross M.M. Shabanowitz J. Hunt D.F. White F.M. Phosphoproteome analysis by mass spectrometry and its application to Saccharomyces cerevisiae.Nat. Biotechnol. 2002; 20: 301-305Google Scholar) have noted that in their experimental setup the multiply phosphorylated peptides were preferentially enriched by IMAC comparing to the singly phosphorylated peptides. That seemed to suggest that the experiments were performed under a condition of limited IMAC capacity. It was proven true by our results in which no preference for multiply phosphorylated peptides was observed. However, the extra capacity of IMAC resin may have contributed to the detection of the nonphosphorylated antibody peptides. Because different pTyr antibodies have slightly different affinity preferences toward different pTyr-containing proteins (22Stancato L.F. Petricoin II, E.F. Fingerprinting of signal transduction pathways using a combination of anti-phosphotyrosine immunoprecipitations and two-dimensional polyacrylamide gel electrophoresis.Electrophoresis. 2001; 22: 2120-2124Google Scholar), the combination of several strains of monoclonal antibodies in immunoprecipitation, as one might have predicted, should offer better coverage than that of a single strain. A combination of two anti-pTyr monoclonal antibodies was employed for immunoprecipitation in this study. Another modification/improvement we made was in sample processing prior to methyl esterification. Dialysis was used for buffer exchange instead of C18 cartridges for de-salting. The use of C18 cartridges presented problems in retaining hydrophilic peptides, including some phosphorylated peptides (data not shown). When MS is used in comparative proteomics, relative quantitation can be achieved through isotope labeling. Considering the small sample size and the multiple steps involved in the work flow, careful control in experiments for reproducibility is required to achieve reliable quantitation. During our experiments, we found that the intensity ratios of hydrogen:deuterium of peptides varied mostly within ± 0.25. Thus, only the pairs with a hydrogen:deuterium ratio change that exceeded 0.5 (50%) were considered as meaningful changes. A tyrosine-phosphorylated peptide of α-tubulin (Tyr-271) was identified to be present in the treated sample but not in the control sample. It is well known that many signaling events lead to modification of microtubules including α-tubulin (for review, see Ref. 23Gundersen G.G. Cook T.A. Microtubules and signal transduction.Curr. Opin. Cell Biol. 1999; 11: 81-94Google Scholar). However, this specific tyrosine phosphorylation site was not previously documented. The detailed relationship between IFNα signaling and the tyrosine phosphorylation of α-tubulin will require further investigation. Two serine-phosphorylated peptides of a novel factor LOC257106 were found to be in the immunoprecipitation eluate of IFNα-treated samples but not in the control samples. The finding of this CDC42 GAP-like protein is of particular interest. Based on the publicly available microarray data, this gene seems most abundantly expressed in T cells (including Jurkat cells) and to a lesser degree in B cells. The induction of gene expression by IFNα was unlikely because the treatment lasted only for 5 min. The peptides identified were serine phosphorylated. However, in an analysis of total phosphorylation, we failed to observe any quantitative changes caused by IFNα treatment, supporting the notion that the treatment did not induce changes in phosphorylation at this particular site. Another possibility was that LOC257106 was tyrosine phosphorylated at a site that was difficult to detect as a tryptic peptide. By sequence analysis, however, no potential tyrosine phosphorylation site was reliably predicted in LOC257106. Alternatively, rather than asserting the effect through direct phosphorylation, LOC257106 might be involved in IFNα signaling pathway through its interaction with a protein of which the tyrosine phosphorylation was affected by IFNα treatment. Scansite (scansite.mit.edu) analysis indicates that besides the CDC42 GAP domain, LOC257106 also possesses several potential SH3 binding domains, including one Grb2 SH3 binding domain (percentile 0.07%), one Cortactin SH3 binding domain (percentile 0.005%), and one Src SH3 binding domain (percentile 0.026%). These SH3 binding domains are known to be involved in signal transduction via protein-protein interaction (24Vidal M. Gigoux V. Garbay C SH2 and SH3 domains as targets for anti-proliferative agents.Crit. Rev. Oncol. Hematol. 2001; 40: 175-186Google Scholar). There are numerous reports indicating the involvement of Rho-type GAP proteins in signal transductions (25Moon S.Y. Zheng Y. Rho GTPase-activating proteins in cell regulation.Trends Cell Biol. 2003; 13: 13-22Google Scholar). However the involvement of this protein in IFNα signaling or any other biological pathways has not been reported previously. We speculate that LOC257106 may be particularly important in signal transduction in immune cells. The exact mode of function requires further investigation. The identification of the involvement of α-tubulin and LOC257106 in IFNα signaling demonstrates that the anti-pTyr immunoprecipitation enriches not only proteins with phosphorylated tyrosine but also their interacting partners. This strategy will broaden our view on tyrosine phosphorylation and protein-protein interaction, the two major mechanisms for receptor-mediated signal transduction.