A Mass Spectrometry-based Proteomic Approach for Identification of Serine/Threonine-phosphorylated Proteins by Enrichment with Phospho-specific Antibodies

Although proteins phosphorylated on tyrosine residues can be enriched by immunoprecipitation with anti-phosphotyrosine antibodies, it has been difficult to identify proteins that are phosphorylated on serine/threonine residues because of lack of immunoprecipitating antibodies. In this report, we describe several antibodies that recognize phosphoserine/phosphothreonine-containing proteins by Western blotting. Importantly, these antibodies can be used to enrich for proteins phosphorylated on serine/threonine residues by immunoprecipitation, as well. Using these antibodies, we have immunoprecipitated proteins from untreated cells or those treated with calyculin A, a serine/threonine phosphatase inhibitor. Mass spectrometry-based analysis of bands from one-dimensional gels that were specifically observed in calyculin A-treated samples resulted in identification of several known serine/threonine-phosphorylated proteins including drebrin 1, α-actinin 4, and filamin-1. We also identified a protein, poly(A)-binding protein 2, which was previously not known to be phosphorylated, in addition to a novel protein without any obvious domains that we designate as Frigg. Frigg is widely expressed and was demonstrated to be a protein kinase A substrate in vitro. We identified several in vivo phosphorylation sites by tandem mass spectrometry using Frigg protein immunoprecipitated from cells. Our method should be applicable as a generic strategy for enrichment and identification of serine/threonine-phosphorylated substrates in signal transduction pathways. Although proteins phosphorylated on tyrosine residues can be enriched by immunoprecipitation with anti-phosphotyrosine antibodies, it has been difficult to identify proteins that are phosphorylated on serine/threonine residues because of lack of immunoprecipitating antibodies. In this report, we describe several antibodies that recognize phosphoserine/phosphothreonine-containing proteins by Western blotting. Importantly, these antibodies can be used to enrich for proteins phosphorylated on serine/threonine residues by immunoprecipitation, as well. Using these antibodies, we have immunoprecipitated proteins from untreated cells or those treated with calyculin A, a serine/threonine phosphatase inhibitor. Mass spectrometry-based analysis of bands from one-dimensional gels that were specifically observed in calyculin A-treated samples resulted in identification of several known serine/threonine-phosphorylated proteins including drebrin 1, α-actinin 4, and filamin-1. We also identified a protein, poly(A)-binding protein 2, which was previously not known to be phosphorylated, in addition to a novel protein without any obvious domains that we designate as Frigg. Frigg is widely expressed and was demonstrated to be a protein kinase A substrate in vitro. We identified several in vivo phosphorylation sites by tandem mass spectrometry using Frigg protein immunoprecipitated from cells. Our method should be applicable as a generic strategy for enrichment and identification of serine/threonine-phosphorylated substrates in signal transduction pathways. Signal transduction events involve transmission and amplification of signals from transmembrane receptors to the nucleus. One major conduit of information is mediated via reversible phosphorylation of proteins (1.Krebs E.G. Historical perspectives on protein phosphorylation and a classification system for protein kinases.Philos. Trans. R. Soc. Lond. B. Biol. Sci. 1983; 302: 3-11Google Scholar, 2.Pawson T. Scott J.D. Signaling through scaffold, anchoring, and adaptor proteins.Science. 1997; 278: 2075-2080Google Scholar, 3.Hunter T. Alexander C.B. Cooper J.A. Protein phosphorylation and growth control.Ciba Found. Symp. 1985; 116: 188-204Google Scholar). In eukaryotes, phosphorylation occurs chiefly on serine, threonine, and tyrosine residues. Phosphorylation on serines and threonines is far more abundant than that on tyrosine residues. Indeed, the ratio of phosphorylation on phosphoserine:phosphothreonine:phosphotyrosine is 1800:200:1 in vertebrates (4.Hunter T. The Croonian Lecture 1997. The phosphorylation of proteins on tyrosine: its role in cell growth and disease.Philos. Trans. R. Soc. Lond. B Biol. Sci. 1998; 353: 583-605Google Scholar). The overall level of phosphorylation is regulated by an interplay between the activities of protein kinases, as well as protein phosphatases within cells. Although tyrosine phosphorylation is far less frequent than serine/threonine phosphorylation in cells, it has been studied quite intensively. One of the major reasons for this is the availability of excellent antibodies that recognize phosphorylated tyrosine residues in immunoprecipitation, as well as Western blotting experiments. In fact, detection of a protein by these antibodies generally constitutes proof that the protein is tyrosine-phosphorylated. Antibodies that specifically recognize serine/threonine-phosphorylated proteins have become available in recent years. However, a major drawback is that they cannot be used for enrichment purposes, because they do not work in immunoprecipitation experiments. Although strategies such as immobilized metal affinity chromatography have been developed for selective enrichment of phosphorylated peptides (5.Posewitz M.C. Tempst P. Immobilized gallium(III) affinity chromatography of phosphopeptides.Anal. Chem. 1999; 71: 2883-2892Google Scholar, 6.Cao P. Stults J.T. Phosphopeptide analysis by on-line immobilized metal-ion affinity chromatography-capillary electrophoresis-electrospray ionization mass spectrometry.J. Chromatogr. A. 1999; 853: 225-235Google Scholar, 7.Stensballe A. Andersen S. Jensen O.N. Characterization of phosphoproteins from electrophoretic gels by nanoscale Fe(III) affinity chromatography with off-line mass spectrometry analysis.Proteomics. 2001; 1: 207-222Google Scholar, 8.Xhou W. Merrick B.A. Khaledi M.G. Tomer K.B. Detection and sequencing of phosphopeptides affinity bound to immobilized metal ion beads by matrix-assisted laser desorption/ionization mass spectrometry.J. Am. Soc. Mass Spectrom. 2000; 11: 273-282Google Scholar), their use to enrich for phosphorylated proteins has not yet been demonstrated. Other strategies to enrich for serine/threonine-phosphorylated proteins have been published, but they involve a series of chemical modifications that can result in loss of proteins (9.Oda Y. Nagasu T. Chait B.T. Enrichment analysis of phosphorylated proteins as a tool for probing the phosphoproteome.Nat. Biotechnol. 2001; 19: 379-382Google Scholar, 10.Goshe M.B. Conrads T.P. Panisko E.A. Angell N.H. Veenstra T.D. Smith R.D. Phosphoprotein isotope-coded affinity tag approach for isolating and quantitating phosphopeptides in proteome-wide analyses.Anal. Chem. 2001; 73: 2578-2586Google Scholar, 11.Zhou H. Watts J.D. Aebersold R. A systematic approach to the analysis of protein phosphorylation.Nat. Biotechnol. 2001; 19: 375-378Google Scholar). In this report, we have tested several antibodies for their ability to immunoprecipitate and Western blot serine/threonine-phosphorylated proteins. We show that some of these antibodies can be used to specifically immunoprecipitate proteins phosphorylated on serine/threonine residues. We performed a large scale purification of proteins whose phosphorylation was induced by treatment of cells with calyculin A, a phosphatase inhibitor. Mass spectrometry has emerged as the technique of choice for identification of proteins, as well as for analysis of phosphorylation (12.Aebersold R. Goodlett D.R. Mass spectrometry in proteomics.Chem. Rev. 2001; 101: 269-295Google Scholar, 13.Resing K.A. Ahn N.G. Applications of mass spectrometry to signal transduction.Prog. Biophys. Mol. Biol. 1999; 71: 501-523Google Scholar). Therefore, we excised bands that were only observed in calyculin A-treated cells from one-dimensional gels and used matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) 1The abbreviations used are: MALDI-TOF, matrix-assisted laser desorption/ionization time-of-flight; LC-MS/MS, liquid chromatography tandem mass spectrometry; MS/MS, tandem mass spectrometry; PKA, protein kinase A; pSer, phosphoserine; pThr, phosphothreonine; PABP, poly(A)-binding protein. 1The abbreviations used are: MALDI-TOF, matrix-assisted laser desorption/ionization time-of-flight; LC-MS/MS, liquid chromatography tandem mass spectrometry; MS/MS, tandem mass spectrometry; PKA, protein kinase A; pSer, phosphoserine; pThr, phosphothreonine; PABP, poly(A)-binding protein. or nanoelectrospray tandem mass spectrometry for identification. Several proteins were identified from this screen including drebrin 1, poly(A)-binding protein 2, myosin heavy chain, γ-actin, and filamin-1. Using antibodies directed against filamin-1, we confirmed that the endogenous protein is phosphorylated on threonine residues, which is consistent with the observation that it has been shown to be a substrate of protein kinase A (PKA) based on in vitro studies. We also identified a novel protein as a phosphoprotein that does not contain any obvious domains from this large scale experiment. We have designated this protein as Frigg, which means “the one who loves,” after the wife of Odin, a Nordic god. Northern blot analysis showed that Frigg is widely expressed. Using an epitope-tagged version of Frigg, we show that it is phosphorylated upon treatment of transfected 293T cells by calyculin A and that it is a substrate of PKA by in vitro phosphorylation experiments. Finally, we identified several serine and threonine phosphorylation sites using immunoprecipitated Frigg protein by tandem mass spectrometry, one of which corresponds to a consensus site for PKA. HeLa and 293T cells were grown in Dulbecco's modified Eagle's medium with 10% fetal bovine serum plus antibiotics. Anti-phosphoserine (7F12, 1C8) and anti-phosphothreonine antibodies (1E11) were from Alexis Biochemicals (San Diego, CA). Anti-phosphoserine (1C8) is a monoclonal mouse IgM antibody that recognizes a broad range of serine-phosphorylated proteins, preferring positively charged amino acids directly neighboring the phosphoserine residues. Anti-phosphoserine (7F12) and anti-phosphothreonine (1E11) are monoclonal mouse IgG antibodies. The 7F12 antibody has the same specificity as 1C8, and 1E11 recognizes a broad range of phosphothreonine-phosphorylated proteins. Anti-pSer/Thr PKA substrate (catalog number 9621), anti-pThr (catalog number 9381), and anti-pThr-Pro (catalog number 9391) antibodies were purchased from Cell Signaling Technology (Beverly, MA). Anti-pSer/Thr and anti-pThr are rabbit polyclonal antibodies that were raised against a synthetic PKA substrate peptide and synthetic phosphothreonine-containing peptides, respectively. Anti-pThr-Pro is a monoclonal mouse IgM antibody generated using synthetic phosphothreonine-proline peptides. Anti-FLAG antibody was from Sigma, and anti-filamin-1 antibody was from Santa Cruz Biotechnology (Santa Cruz, CA). Metabolic labeling of HeLa and 293T cells was done with ∼107 cells in Dulbecco's modified Eagle's labeling medium. 0.2 mCi/ml of 35S-labeled cysteine and methionine (Translabel; ICN) was added and incubated for 6–12 h at 37 °C. After metabolic labeling, the cells were treated with 0.1 μg/ml calyculin A for 30 min at 37 °C or left untreated. The cells were lysed in lysis buffer (50 mm Tris, pH 7.6, 150 mm NaCl, 1% Nonidet P-40, 1 mm sodium orthovanadate, 10 mm sodium fluoride, 5 mm sodium pyrophosphate, and 10 mm β-glycerophosphate) in the presence of protease inhibitors. The lysates were centrifuged for 30 min at 4 °C at 14,000 rpm. Cleared cell lysates were incubated overnight at 4 °C with 20 μl of polyclonal antibody (anti-pThr and anti-pSer/Thr) and 30 μl of protein A-agarose beads or 20 μl of monoclonal antibody (anti-pThr-Pro) and 30 μl of protein l-agarose beads. The monoclonal antibodies from Alexis (7E12, 1C8, and 1E11) were first redissolved and subsequently incubated with cleared cell lysates and 30 μl of protein L agarose beads. Precipitated immune were in lysis and the bound proteins were by for min in The proteins were by and to Western blotting cells were used but treated as After the was The were in in 20 overnight at 4 °C and incubated with the The were and incubated with antibody and developed using detection to the HeLa cells grown to in were overnight in Dulbecco's modified Eagle's medium. The medium was to medium 1 h to The cells were subsequently for h with One was treated with calyculin A for 30 from untreated and treated were lysed in lysis buffer protease and cleared by at in a of the cleared was done with 20 μl of antibody with 30 μl of protein A-agarose for h at 4 °C. The were in lysis in and to The was and the bound was by the large scale shown in HeLa cells were grown and for The cells were treated with 0.1 μg/ml calyculin A for 30 min at 37 °C or left untreated. The cells were lysed in 20 of lysis Cleared cell lysates were incubated with μl of antibody and μl of protein A-agarose beads overnight at 4 °C. Precipitated immune were in lysis and bound proteins were by in The proteins were by and the was by a method with mass analysis and of mass spectrometry to pathways. Scholar). and cell Northern were purchased from and to the A to the protein was used to a epitope-tagged The was by a using the and The was with and and 293T cells were transfected with of Frigg or using the h and h after the medium was for h after the cells were for The cells were treated with 0.1 μg/ml calyculin A for 30 min at 37 °C or left untreated. were lysed and to immunoprecipitation and Western blotting as with Frigg, cells were lysed at 4 °C in lysis Cleared cell lysates were immunoprecipitated with 10 of antibody for h at 4 °C. After the beads were in lysis buffer and in kinase buffer (50 mm pH 10 mm 5 mm were incubated for 30 min at 30 °C in 5 μl of mm and 20 of and of protein kinase A Protein samples were by in buffer and by The was and on a Mass identification of proteins was done to a strategy previously O.N. P. Mass identification and of proteins from electrophoretic strategies and 1998; Scholar). The to the bands in were After and the proteins were with an of modified sequencing The was overnight at 37 °C. After the was to with and a nanoscale and to R. P. purification and technique based on for the analysis of peptide by matrix-assisted laser desorption/ionization mass spectrometry.J. Mass Spectrom. 1999; Scholar). peptides were with 1 μl of in The was as a series of a mass of peptide mass was performed using and The samples for analysis were as the peptides were with directly a nanoelectrospray tandem mass spectrometry was performed on a liquid mass analysis was performed using a mass A liquid chromatography system The was used to a of samples were a using using a was using a of in an of analysis was performed using and and the was using the enrich serine/threonine-phosphorylated proteins, we to first with several available antibodies. HeLa cells were with and left untreated or treated with calyculin A. with calyculin A the phosphorylation of several protein substrates on serine/threonine residues because of the of protein within cells. Cell lysates were immunoprecipitated with several antibodies that were directed against phosphorylated proteins. In we that of such antibodies well in Western blotting they are generally not in immunoprecipitating proteins and cannot be used for enrichment shown in of these antibodies not immunoprecipitate any bands that were between untreated and calyculin A-treated cells. However, of the antibodies showed that that they were specifically to immunoprecipitate proteins from calyculin A-treated cells antibodies were raised against residues, that it recognizes in the of a or a of phosphoserine and residues, and are available We to of the antibodies that well in immunoprecipitation experiments. this cell lysates from HeLa cells were to Western blotting with anti-pThr and antibodies. that these antibodies were to proteins whose phosphorylation was induced by treatment of cells with calyculin A treatment in Western blotting experiments. The of the bands observed was to that in the immunoprecipitation in or Therefore, these antibodies can be used for immunoprecipitation, as well as Western blotting experiments. A large scale was performed using HeLa cells that were and One was left and the was treated with calyculin A for 30 Cell lysates were immunoprecipitated using The immune were and by one-dimensional and the was to proteins major protein bands that were observed specifically in the calyculin A-treated cells were by and by mass shown in 1 was identified as by mass is also known as protein and is an of the it of in to a of and A. in phosphorylation with protein kinase and by of a of Biophys. 1997; Scholar). In has been shown to with to and to to receptors Phosphorylation of the to the and Biol. 2001; Scholar, protein the of and in the Cell Biol. 1997; Scholar, of by protein 2000; Scholar). is a known phosphoprotein with serine as a site for PKA A. in phosphorylation with protein kinase and by of a of Biophys. 1997; Scholar, Phosphorylation of the to the and Biol. 2001; Scholar, of a protein kinase phosphorylation site in the of Biophys. 2000; Scholar). We to is also phosphorylated on threonine residues upon treatment of HeLa cells with calyculin A. was immunoprecipitated with anti-filamin-1 antibody by Western blotting with anti-pThr phosphorylation on threonine residues upon calyculin A treatment that it is phosphorylated on threonine, in addition to serine, residues is phosphorylated by calyculin A HeLa cells were left untreated or treated with 0.1 μg/ml calyculin A for 30 Cleared cell lysates were immunoprecipitated with anti-filamin-1 and Western with anti-pThr The a Western blot using anti-pThr antibody to immunoprecipitated from untreated or calyculin A-treated cells. The is a Western blot from a to show of was identified as myosin heavy A heavy is a myosin in to the and has been identified in a range of cells including and cells T. R. to the and Scholar). The of myosin is its to and in because of the of The of myosin is known to be regulated by its phosphorylation T. myosin a role of its Biol. 2001; Scholar). In for phosphorylation of threonine residues and is which myosin from to a phosphorylation of threonine by the myosin heavy kinase Biol. Chem. 2001; Scholar). mass spectrometry was performed to the of the protein at in was to be drebrin 1 not is an protein identified in as a regulated protein expressed in T. proteins with in 1985; Scholar). have been identified in the T. Y. of a for the regulated protein, and T. of drebrin A and of in Scholar, T. S. S. of drebrin and of its Biophys. that are generated by of a T. R. H. of of by drebrin A expressed in Cell Scholar). In addition to its ability to with drebrin 1 the of and α-actinin R. R. T. A. T. analysis of the and activities of Cell 1999; Scholar, Y. T. R. by drebrin of the of a protein in of growth Scholar, R. T. Y. T. Y. a protein from the of from Biol. Chem. Scholar). R. R. T. A. T. analysis of the and activities of Cell 1999; have shown that it is phosphorylated in peptides from 4 were from α-actinin 4 a protein as a of the in T. R. to the and Scholar). is of an an and a H. H. H. Y. of with Biol. Chem. 1997; Scholar). is a of of proteins, which to the and is to the to between and the 1 in Cell Biol. Scholar, of the site within the 1 Biol. Chem. Scholar). of α-actinin have been identified 1 to and α-actinin 4 is to be in cell and in H. H. H. Y. of with Biol. Chem. 1997; have shown that α-actinin 4 is a substrate for protein kinase which it in the In it has been shown to be by in on P. B. B. phosphorylation of in Biol. Chem. 1999; Scholar, B. The of is phosphorylated on its by the Biol. Chem. 2001; Scholar). The protein in was identified by as poly(A)-binding protein is an protein that with affinity to their and their A. the for of poly(A)-binding protein 2000; Scholar). of the in the to B. S. P. A. in the 1998; Scholar). has been demonstrated to with and to a role in signal transduction S. T. T. The of an poly(A)-binding protein 2, with and Mol. 2001; Scholar). H. D.R. The phosphorylation of poly(A)-binding protein its to and its with and Biol. Chem. 2000; have shown that the level of phosphorylation of PABP, a protein to the of to and to proteins we enriched with an we that it is a serine-phosphorylated protein, and its level of phosphorylation have as is the with the of proteins that were identified from the shown in The peptides from the protein at in 5 a novel protein that corresponds to a that was to a protein of phosphoproteins identified by mass on serine A. in phosphorylation with protein kinase and by of a of Biophys. 1997; Scholar, Phosphorylation of the to the and Biol. 2001; Scholar, of by protein 2000; Scholar, of a protein kinase phosphorylation site in the of Biophys. 2000; heavy on threonine residues and in T. myosin a role of its Biol. 2001; Scholar, phosphorylation of threonine by the myosin heavy kinase Biol. Chem. 2001; in T. of drebrin A and of in Scholar, R. R. T. A. T. analysis of the and activities of Cell 1999; in by protein kinase in vitro by on H. H. H. Y. of with Biol. Chem. 1997; Scholar, of the site within the 1 Biol. Chem. B. The of is phosphorylated on its by the Biol. Chem. 2001; on serine and sites of PABP, a protein to the of to and to A. the for of poly(A)-binding protein 2000; H. D.R. The phosphorylation of poly(A)-binding protein its to and its with and Biol. Chem. 2000; upon of S. H. Jensen O.N. a of phosphorylation in Biophys. 1997; in a the protein by was we to it was in the this been with a of analysis was used for of the by the from this against S. A. A of the in 2001; Scholar). We that the of amino at the methionine at that this was the The of this a protein of amino acids with a mass of protein was designated as Frigg after a who was the wife of in Nordic of the protein using such as and not the presence of any P. a identification of Sci. S. A. 1998; Scholar, A. R. and of protein 1998; Scholar). A against the protein showed that this protein was to which also not any obvious the of Frigg, a Northern blot analysis was shown in it is expressed as a of in and cell Frigg at the level of the protein, the was by and a epitope-tagged The was transfected 293T cells by the and the expressed protein was immunoprecipitated with The immune were to one-dimensional and Western with to The Western blot that the expressed protein at which is consistent with of the protein observed in the large scale purification experiment. The phosphorylation of epitope-tagged Frigg was by transfected 293T cells with calyculin A. After enrichment by immunoprecipitation with the proteins were by one-dimensional and Western with anti-phosphothreonine The Western blot that Frigg is phosphorylated in vivo as a result of calyculin A treatment The antibody used in the large scale immunoprecipitation recognizes with an in the is the consensus for protein kinase A. in vitro phosphorylation was performed to Frigg was a substrate of PKA. Frigg was incubated with in the or presence of the of PKA for 30 The proteins were by in buffer and subsequently by The result of this in vitro phosphorylation was on a The that Frigg phosphorylation incubated with the of PKA that Frigg is a substrate of PKA the of phosphorylation in such in vitro phosphorylation can be we used Frigg protein that was immunoprecipitated from calyculin A-treated cells to identify in vivo phosphorylation The of such an is because in vitro phosphorylation can were to serine/threonine phosphorylation in Frigg. peptides were in a of from 5 to and by tandem mass spectrometry on a time-of-flight mass peptides from Frigg were of these were identified as phosphoserine/phosphothreonine-containing peptides one or phosphorylation sites by the and on the residues were identified as phosphorylation sites in of the peptides Phosphorylation on serine residues was confirmed by the detection of in the to loss of by Phosphorylation on serine residues was on residues and The the presence of several phosphorylation sites in Frigg, but the lack of of the ion in some detection of phosphorylation a of such phosphopeptides that were identified as phosphorylated the site of phosphorylation not be of peptides in Frigg phosphorylation sites were by mass amino in a of phosphorylated peptides phosphorylation site not be by mass of the number of on peptides identified by the number of on peptides identified by in a of phosphorylation sites on Frigg by mass analysis of peptides from Frigg. phosphoserine/phosphothreonine-containing peptides were identified in the phosphorylation sites were identified in of the The of the peptide is and are for with an were generated from the peptide in which the phosphoserine been to by of A, of the of a charged peptide ion at the of loss of from the of an of a charged peptide The phosphorylation site was to the serine in in the peptide in The serine at is of a consensus that is phosphorylated by protein kinase A. of an of a charged peptide The phosphorylation site was to the serine in 5 of the peptide in phosphorylation is an in a number of signal transduction it is to the and for a a major for serine/threonine-phosphorylated proteins was the lack of availability of immunoprecipitating antibodies. this a number of strategies based on chemical of phosphoserine and residues have been A drawback of these is that they a series of chemical and purification that to loss of phosphorylated proteins. In this report, we have shown that antibodies directed against phosphoserine and residues can be used for enrichment of these proteins for a analysis as we shown for tyrosine phosphorylation A. B. of by mass identification of as a substrate of the and growth Sci. S. A. 2000; Scholar). phosphorylation not several of the phosphorylation sites and phosphopeptides that we have of is to we have previously in the of tyrosine phosphorylation site substrates in the growth H. B. A. phosphorylation of the growth Biol. Chem. Scholar).