Tagging-via-Substrate Strategy for Probing O-GlcNAc Modified

May 20, 2005 - Interpreting the protein language using proteomics. Ole N. Jensen. Nature Reviews Molecular Cell Biology 2006 7 (6), 391-403 ...
1 downloads 0 Views 242KB Size
Download PDF
Tagging-via-Substrate Strategy for Probing O-GlcNAc Modified Proteins Robert Sprung,† Animesh Nandi,† Yue Chen,† Sung Chan Kim,† Deb Barma,† John R. Falck,†,‡ and Yingming Zhao*,†,‡, Departments of Biochemistry and Pharmacology, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas 75390-9038 Received February 14, 2005

Identification of proteins bearing a specific post-translational modification would imply functions of the modification. Proteomic analysis of post-translationally modified proteins is usually challenging due to high complexity and wide dynamic range, as well as unavailability of efficient methods to enrich the proteins of interest. Here, we report a strategy for the detection, isolation, and profiling of O-linked N-acetylglucosamine (O-GlcNAc) modified proteins, which involves three steps: metabolic labeling of cells with an unnatural GlcNAc analogue, peracetylated azido-GlcNAc; chemoselective conjugation of azido-GlcNAc modified proteins via the Staudinger ligation, which is specific between phosphine and azide, using a biotinylated phosphine capture reagent; and detection and affinity purification of the resulting conjugated O-GlcNAc modified proteins. Since the approach relies on a tag (azide) in the substrate, we designated it the tagging-via-substrate (TAS) strategy. A similar strategy was used previously for protein farnesylation, phosphorylation, and sumoylation. Using this approach, we were able to specifically label and subsequently detect azido-GlcNAc modified proteins from the cytosolic lysates of HeLa, 3T3, COS-1, and S2 cell lines, suggesting the azido-substrate could be tolerated by the enzymatic systems among these cells from diverse biological species. We isolated azido-GlcNAc modified proteins from the cytosolic extract of S2 cells and identified 10 previously reported and 41 putative O-GlcNAc modified proteins, by nano-HPLC-MS/MS. Our study demonstrates that the TAS approach is a useful tool for the detection and proteomic analysis of O-GlcNAc modified proteins. Keywords: glycosylation • O-GlcNAc • proteomics • tagging-via-substrate • Staudinger ligation • post-translational modifications

Introduction A major goal of proteomics studies is the global analysis of the interactions, expression, and modifications of the diverse and dynamic cellular proteins and to correlate this information with protein function and cellular regulation. Unfortunately, due to the wide dynamic range and high complexity of a cell’s proteome, the global profiling of low-to-medium abundance proteins and post-translational modifications is impractical. More powerful separation and mass spectrometry tools are necessary to deal with this daunting challenge. One strategy to address the problems is to develop robust technologies for the selective detection and isolation of proteins bearing specific modifications. The O-GlcNAc modification, an abundant post-translational modification present at serine and threonine residues of nuclear and cytosolic proteins, was discovered by Hart and his colleagues about two decades ago.1 The modification is found in all higher eukaryotes from C. elegans to mammals and is * To whom correspondence should be addressed. Tel: (214) 648-7947. Fax: (214) 648-2797. E-mail: [email protected]. † Departments of Biochemistry. ‡ Pharmacology.

950

Journal of Proteome Research 2005, 4, 950-957

Published on Web 05/20/2005

dynamic, possessing the ability to respond to external stimuli.2-6 The modification has been shown to be involved in the modulation of protein-protein interactions, protein-DNA binding, protein localizations, and protein stability.3 The fast turnover of O-GlcNAc is the result of the interplay between O-GlcNAc transferase (OGT),7 which adds GlcNAc to proteins, and the O-GlcNAc-specific β-N-acetylglucosaminidase (OGlcNAcase),8 which removes it. Both enzymes are ubiquitously expressed in mammalian cells and OGT has been shown to be essential for cell viability.9 An important consideration with regard to the O-GlcNAc modification is that the intracellular level of UDP-GlcNAc, the substrate for GlcNAc transfer reaction, is regulated by glucose. It is estimated that between 2 and 5% of the glucose taken up by a cell is diverted through the hexosamine biosynthetic pathway leading to formation of UDP-GlcNAc.10 Thus, the position of UDP-GlcNAc in the hexosamine biosynthetic pathway, coupled with the rapid turnover of O-GlcNAc, implicates the O-GlcNAc modification as an energy sensor and regulatory modification.3 Indeed, dysregulation of the OGlcNAc modification has been shown to be involved in the development of insulin resistance and diabetes.11-13 10.1021/pr050033j CCC: $30.25

 2005 American Chemical Society

research articles

TAS for O-GlcNAc Modification

Despite being implicated in myriad processes and pathways, the molecular mechanisms by which O-GlcNAc can potentiate these effects remain unknown. Global identification of the dynamically modified O-GlcNAc proteins would help determine how O-GlcNAc exerts its regulatory effects. Unfortunately, as with most posttranslational modifications, there is not an efficient technology available for the selective isolation of O-GlcNAc modified proteins. As a result, the extent of OGlcNAc modification of proteins is largely unknown. Development of an efficient strategy for the enrichment of O-GlcNAc modified proteins will aid in the elucidation of the dynamic patterns and consequences of the modifications that occur in diverse cellular environments. Chemical reactions that are selective and specific to a single chemical moiety in the context of complex biological milieu represent powerful tools for chemical analysis of biologically interesting molecules. An example of such a reaction is the enhanced Staudinger ligation pioneered by the Bertozzi group as a bioorthoganal reaction between a phosphine and an azide.14 The reaction has unique features suitable for biological applications. For example, the reaction can be carried out at room temperature in aqueous solution; there is no crossreaction with endogenous cellular compounds at a significant reaction rate during the experiment time window; and neither phosphines nor azides occur in any known biomolecules. In addition, the phosphine can be engineered to incorporate a wide variety of tags including fluorescent probes and affinity tags such as biotin and the FLAG tag.15-17 The cellular metabolism of azide-tagged sugars has been characterized by the Bertozzi group as it applies to the O-GlcNAc monosaccharide and sialic acid oligosaccharide modifications.18-20 This chemistry has even been demonstrated to be a viable strategy for the labeling of cells in vivo, as indicated by a study involving the use of azide-tagged N-acetyl mannosamine for the detection of sialic acid glycosylation of cell surface proteins in living mice.16 Recently, our lab has applied this improved Staudinger ligation to detection and proteomics of protein farnesylation.17 Briefly, an azide-modified version of farnesylation substrate, azido-farnesyl pyrophosphate, is used to metabolically label cells. The resulting azido-farnesylated proteins are chemoselectively conjugated to a biotinylated phosphine via the Staudinger ligation. The biotinylated, azidofarnesylated proteins are affinity-purified by streptavidin-conjugated agarose beads. LC-MS/MS analysis of the isolated proteins led to the identification of 18 proteins containing the C-terminal CAAX box including a putative farnesylated protein with a unique recognition sequence. We designate the strategy for detection and isolation of post-translationally modified proteins based on a tag introduced in the modification substrate, the tagging-viasubstrate (TAS) technology.17 We have applied the strategy to protein sumoylation (using a HA tag in the sumoylation substrate),21 and protein phosphorylation (using the sulfur atom in ATP-γ-S as a tag).22 Here, we extend the TAS strategy to the detection and identification of O-GlcNAc modified proteins using peracetylated azido-GlcNAc as a substrate for O-GlcNAc modification. We demonstrate that the substrate could be incorporated into proteins for O-GlcNAc modification in the cell lines derived from diverse organisms including human, monkey, mouse and Drosophila, suggesting the enzymatic systems in these cell lines are compatible with the azido modification in the substrate. Our initial experiment on affinity purification of azido-GlcNAc

modified proteins from metabolically labeled S2 cell lysate led to the putative identification of 51 O-GlcNAc modified proteins, including 10 known O-GlcNAc modified proteins. Our results suggest that TAS technology is a useful tool for detection and proteomics of O-GlcNAc modified proteins in order to elucidate the functional consequences of the modification.

Experimental Section Materials. HPLC grade acetonitrile, water, and methanol were from EM Science (Gibbstown, NJ); trifluoroacetic acid (TFA) was from Fluka (Buchs, Switzerland); acetic acid was from Aldrich (St. Louis, MO); ammonium bicarbonate was from Fisher (Fair Lawn, NJ); trypsin was from Promega (Madison, WI). Peracetylated N-(2-azidoacetyl)-glucosamine (peracetylated azido-GlcNAc) was synthesized in-house according to the procedure described previously.15 The synthetic procedures for azido-farnesyl pyrophosphate and biotinylated phosphine capture reagent were the same as previously reported.17 DMEM and Drosophila SFM cell media were purchased from Gibco (Grand Island, NY). Protease inhibitor cocktail was from Roche Molecular Biochemicals (Indianapolis, IN). Ammonium bicarbonate, SDS and salts used in buffer preparation were from Fisher Scientific (Fair Lawn, NJ). BSA, glucosamine, phosphate buffered saline, Tween-20, and urea were purchased from Sigma (St. Louis, MO). Streptavidin polyacrylamide beads were from Pierce (Rockford, IL). Nitrocellulose membrane and 4-20% gradient minigels were obtained from Bio-Rad Laboratories (Hercules, CA). Western Lightning Chemiluminescence Reagent Plus was from Perkin-Elmer Life Science (Boston, MA). Cell Culture. HeLa, 3T3, and Cos-1 Cells. Dishes (15 cm) of HeLa, 3T3 and Cos-1 cells were grown to 70% confluence in complete DMEM (4.5 g glucose/L) with 10% FBS and antibiotics in a 37 °C, 5% CO2 incubator. For labeling, media was aspirated and the cells were washed with PBS. Fifteen mL DMEM containing 250 µM peracetylated azido-GlcNAc and 1 g glucose/L was then added to each dish and cells were incubated for 24 h. S2 Cells. Dishes (15 cm) of S2 cells were grown to confluence in Drosophila SFM (2 g glucose/L, 2 g trehalose/L) with 10% FBS and antibiotics at 25 °C. For labeling, media was aspirated and the cells were washed with PBS. Cells were labeled for 24 h with 250 µM peracetylated azido-GlcNAc in Drosophila SFM diluted 1:4 with S2 chemical-defined media containing no sugars (10% FBS, 0.5 g glucose/L, 0.5 g trehalose/L final, see Supporting Information for formulation). Preparation of Cytosolic Extract. Three hours prior to harvest, 150 µL of 400 mM glucosamine (4 mM final) was added to plates to inhibit O-GlcNAcase, the enzyme responsible for the removal of the O-GlcNAc modification, during the harvesting and lysis procedures. Each plate of cells was harvested by scraping in 2 mL PBS containing 4 mM glucosamine and pelleted by centrifugation at 1000 × g and 4 °C for 10 min. The cell pellet was resuspended in an equal volume of icecold hypotonic lysis buffer (10 mM KCl, 1.5 mM MgCl2, 1 M Hepes, pH 7.4) with protease inhibitors and 4 mM glucosamine, incubated on ice for 30 min and homogenized using 20 passes in a Dounce homogenizer employing a B-rated pestle. The lysate was subjected to ultracentrifugation at 100 000 × g for 1 h at 4 °C to remove intact cells, cell membranes and organelles including nuclei. The supernatant was collected and considered the cytosolic extract. Journal of Proteome Research • Vol. 4, No. 3, 2005 951

research articles Biotin-Conjugation Reaction. Protein was precipitated from the cytosolic extract by acetone precipitation at -20 °C overnight. Precipitated proteins were pelleted by centrifugation at 15 000 × g for 10 min, washed twice with 1 mL cold acetone and redissolved in PBS containing 1% SDS for a final concentration of 4 mg/mL. To specifically biotinylate azido-GlcNAc modified proteins, a Staudinger capture reaction using biotinylated phosphine capture reagent (bPPCR) was performed on cytosolic lysates. For HeLa, 3T3, and Cos-1 cell lysates, a 100 µL solution consisting of 150 µg protein and 50 µM bPPCR in 1% SDS was reacted at room-temperature overnight with shaking in the presence or absence of 100 µM azido-farnesyl pyrophosphate. For S2 cell lysate, a 100 µL solution consisting of 150 µg protein and 500 µM bPPCR in 1% SDS was reacted at room-temperature overnight with shaking in the presence or absence of 1 mM azido-farnesyl pyrophosphate. Western Blot Analysis. Biotin-conjugated samples were precipitated in 9 volumes ice-cold acetone at -20 °C for >2 h to remove excess bPPCR and resuspended in SDS-PAGE sample buffer. Twenty-µg samples were run on a 4-20% gradient minigel. Proteins were then transferred to a nitrocellulose membrane. The membrane was blocked in 5% BSA TBS (25 mM tris, pH 7.5, 150 mM NaCl) with shaking at room temp for 1 h. The membrane was probed for 1 h with a 1:20 000 dilution of streptavidin-conjugated HRP in the presence and absence of 100 µM D-biotin. After probing, the membrane was washed 6 times with shaking for 15 min each time in TBST (25 mM tris, pH 7.5, 150 mM NaCl, 0.4% Tween-20). The blot was developed using the Western Lightning chemiluminescence reagent. Affinity Purification of Biotinylated, O-GlcNAc Modified Proteins. Excess bPPCR was removed from biotin-conjugated samples by precipitating proteins in 9 volumes ice-cold acetone at -20 °C for >2 h. Protein pellets were resuspended in 1% SDS/PBS. Ten mg protein was added to a 20 µL volume of streptavidin-conjugated polyacrylamide beads. Samples were rocked at 4 °C for 2 h. To remove the beads, samples were centrifuged at 1000 × g for 5 min and the supernatant decanted. The beads were then washed three times each using 8 M urea, 1% SDS/PBS and 50 mM NH4CO3 (pH 8.0). Nano-HPLC/Mass Spectrometry for Protein Identification. After affinity purification, 50 ng trypsin was added directly to the beads. The digestion was carried out at 37 °C overnight. The tryptic peptides were sequentially extracted with 50% acetonitrile/50% water (v/v), and 75% acetonitrile/25% water (v/v) solutions. The peptide extracts were combined and dried in a SpeedVac. The peptide samples were cleaned with ZipTip C18 (Millipore, Bedford, MA) prior to nano-HPLC/tandem mass spectrometry analysis. Nano-HPLC/tandem mass spectrometry analysis was performed in a LCQ DECA XP ion-trap mass spectrometer (ThermoFinnigan, SanJose, CA) equipped with a nano ESI source (ThermoFinnigan, SanJose, CA). The electrospray source was coupled online with an Agilent 1100 series nano flow HPLC system (Agilent, Palo Alto, CA). Two µL of the peptide solution in buffer A (2% acetonitrile/97.9% water/0.1% acetic acid (v/ v/v)) was manually loaded into a capillary HPLC column (50 mm length × 75 µm ID, 5 µm particle size, 300 Å pore diameter) packed in-house with Luna C18 resin (Phenomenex, St. Torrance, CA). The peptides were eluted from the column with a gradient of 5% to 80% buffer B (90% acetonitrile/9.9% water/ 0.1% acetic acid (v/v/v)) in buffer A over 30 min. The eluted 952

Journal of Proteome Research • Vol. 4, No. 3, 2005

Sprung et al.

peptides were electrosprayed directly into the LCQ mass spectrometer. The MS/MS spectra were acquired in a data-dependent mode that determined the masses of the parent ions and fragments of the three strongest ions. All MS/MS spectra were searched against the NCBI nonredundant protein sequence database using the MASCOT algorithm (Matrix Solution, London, England) for protein identification. The following search parameters were used in all MASCOT searches: maximum of one missed trypsin cleavage and a maximum 3-Da error tolerance in the MS and 0.8-Da in MS/MS data. Those protein hits with three ions matched, each with a score higher than 35, were considered true protein identifications. All other hits were manually analyzed to ascertain the accuracy of protein identification. In the manual analysis, the criteria used for a true identification was that the masses of all the major peaks (typically more than 7 peaks) in a MS/MS spectrum had to match those of the theoretically calculated fragment ions and the mass errors were within 0.5 Da.

Results Strategy. Our strategy for detection and isolation of OGlcNAc modified proteins involves three steps: (1) the metabolic labeling of a cell with peracetylated azido-GlcNAc. Peracetylated azido-GlcNAc would be expected to penetrate the cell surface membrane due to the hydrophobic nature of the modified sugar and be processed by a series of enzymatic reactions inside the cells, leading to the formation of UDPazido-GlcNAc (Figure 1), the substrate for the O-GlcNAc modification. The resulting O-GlcNAc modified proteins would contain an azide moiety. (2) The azido-GlcNAc modified proteins are extracted and chemoselectively conjugated to a biotinylated phosphine capture reagent via the Staudinger ligation. Thus, a biotin handle is attached to the azide tag in the O-GlcNAc modified proteins. And (3) the biotinylated, O-GlcNAc modified proteins are detected and/or isolated for protein identification. A similar strategy for the detection of O-GlcNAc modified proteins has been described by the Bertozzi group,15 where FLAG tag-containing phosphine has been used to conjugate azido-GlcNAc modified proteins for subsequent detection. In the present study, a biotinylated phosphine is used as the conjugation reagent. The biotin moiety offers the advantages of efficient detection and affinity purification due to the strong interaction between biotin and streptavidin. The biotin-streptavidin interaction can withstand harsh washing conditions, such as 8 M urea and 1-2% SDS, allowing the removal of nonspecifically bound proteins. Previous strategies using the FLAG peptide tag and antibody detection are not amenable to such stringent wash steps. Specific Detection of O-GlcNAc Modified Proteins from azido-GlcNAc labeled Cells by Western Blot Analysis. To test if peracetylated azido-GlcNAc can metabolically label proteins for detection, cells were incubated with either peracetylated GlcNAc (control) or peracetylated azido-GlcNAc. The cytosolic proteins were extracted, and a capture reaction was performed to specifically biotinylate the azido-GlcNAc labeled proteins using biotinylated phosphine capture reagent. The proteins were then precipitated to remove excess phosphine and analyzed by Western blotting method using streptavidin-HRP (Figure 2). For HeLa, 3T3, COS-1, and S2 cell lines, the signals were detected for cells labeled with peracetylated azido-GlcNAc, but not with peracetylated GlcNAc. The signals were dependent

TAS for O-GlcNAc Modification

research articles

Figure 1. TAS strategy for the global detection of O-GlcNAc modified proteins. (A) Peracetylated N-(2-azidoacetyl)-glucosamine (1) is delivered into the cells through metabolic labeling and converted to UDP-azido-GlcNAc (3), which is used by O-GlcNAc transferase for O-GlcNAc modification of proteins (4). (B) Biotin-conjugation reaction using biotinylated phosphine capture reagent (5) results in the selective biotinylation of azido-GlcNAc modified proteins (6).

on the capture reaction step and could be competed out by performing the capture reaction in the presence of exogenous azido compound (azidofarensyl pyrophosphate) or by probing the nitrocellulose membranes with streptavidin-HRP in the presence of exogenous biotin. These results suggest that the signals detected are specific to azido-GlcNAc. Since O-GlcNAc is the major form of glycosylation for cytosolic proteins, these results indicate that cells can incorporate the exogenous azidoGlcNAc substrate into proteins for O-GlcNAc modification. Our experiment suggests that TAS technology can be successfully applied to O-GlcNAc studies of diverse organisms including human, mouse, monkey and Drosophila. Capillary-HPLC/Mass Spectrometric Analysis of Affinity Purified Cytosolic Proteins. Only about 80 O-GlcNAc modified proteins have been previously identified by various methods.3 None of the previous work addressed O-GlcNAc modified proteins in Drosophila cells. To identify O-GlcNAc modified proteins in Drosophila cells, 10 mg cytosolic proteins of Drosophila S2 cells labeled with either peracetylated GlcNAc or peracetylated azido-GlcNAc were subjected to a capture reaction using biotinylated phosphine capture reagent. The resultant biotinylated proteins were affinity-purified using streptavidin-conjugated beads. Nonspecifically bound proteins were removed by stringent wash conditions (8 M urea, 1% SDS/

PBS). Proteins remaining bound to the beads were digested with trypsin. Tryptic peptides were analyzed by LC-MS/MS and automatic protein sequence database search using MASCOT algorithm for protein identification. Putative O-GlcNAc modified proteins were identified by subtractive analysis of the nonspecifically binding proteins of the control sample from those proteins identified in azidoGlcNAc labeled cell lysates. The analysis led to the identification of 51 proteins, 10 of which have been previously identified as bearing the O-GlcNAc modification (Table 1) and serve as a good positive control. Also, the number of proteins identified here represents an increase in the number of O-GlcNAc modified proteins previously identified in a single study.3 Summary of Identified Proteins. The identified proteins were manually classified according to their annotated functions described in the NCBI (www.ncbi.nlm.nih.gov) and FlyBase (http://flybase.bio.indiana.edu) protein databases (Figure 4). The O-GlcNAc modification has been suggested as a potential sensor of glucose availability and cellular nutritional status.10,23-27 The peracetylated azido-GlcNAc used in this study would be expected to enter the hexosamine biosynthetic pathway at a point after the rate-limiting step of GlcNAc biosynthesis, similar to glucosamine.10 Identification of three proteins involved in glucose metabolism suggests that the Journal of Proteome Research • Vol. 4, No. 3, 2005 953

research articles

Figure 2. Detection of metabolically labeled O-GlcNAc-N3 modified proteins in diverse organisms. Cytosolic proteins from cells in the presence and absence of metabolic labeling with GlcNAcN3 were conjugated to biotinylated phosphine capture reagent (5). The resulting biotinylated proteins were separated by SDSPAGE and detected by Western Blot Analysis, probing with HRPconjugated streptavidin.

Sprung et al.

Figure 3. Example of nano-HPLC-MS/MS analysis for protein identification. (A) Total ion current chromatogram of a nanoHPLC-MS/MS of the beads digest. (B) MS spectrum at retention time of 30.00 min. (C) MS/MS spectrum of 661.5 m/z, which identified the peptide ALIAAQYSGAQVK, unique to translation elongation factor 1 γ.

O-GlcNAc modification might provide feedback regulation in order to adjust these enzymes’ activities in response to energy status. It has been estimated that as much as 25% of cellular energy is used for protein synthesis and processing in cells.28 Thus, it would be reasonable to speculate that the regulation of the transcription/translation machinery of the cell would be coupled to energy availability. The O-GlcNAc modification, a good candidate for a regulatory modification related to glucose availability, has previously been detected on transcription and translation factors.3 Using the TAS technology, we were able to identify 4 transcription and translation factors, 3 of which have been previously reported to be O-GlcNAc modified. O-GlcNAc has also been suggested to play a role in the cellular stress response pathways.29 In support of this notion, 9 (18%) of the proteins identified in our analysis are annotated as being involved in stress response pathways. These include 4 heat shock proteins and proteins involved in the regulation of cellular redox homeostasis. In addition to proteins involved in protein synthesis/ processing, stress response, and metabolism, we also identified 30 other proteins including structural proteins, signaling proteins, cell cycle regulators, and proteins involved in nucleotide metabolism and carbohydrate metabolism. The functional role of the O-GlcNAc modification remains to be established for these proteins. 954

Journal of Proteome Research • Vol. 4, No. 3, 2005

Figure 4. Pie diagram showing functional classes for identified proteins. Classifications are based on NCBI and FlyBase annotations.

Discussion Our goal during the course of this investigation was to explore the TAS technology as an approach for the global detection and identification of O-GlcNAc modified proteins. In addition, we wanted to test whether the enzymatic machinery of cells from diverse organisms would tolerate the azide tag. To achieve these goals, we used peracetylated azidoGlcNAc as a substrate for O-GlcNAc modification. Our experiment demonstrates that peracetylated azido-GlcNAc could be used by the cell lines from diverse biological species including human, mouse, monkey, and Drosophila. These results are in agreement with the extensive characterization previously per-

research articles

TAS for O-GlcNAc Modification Table 1. List of Identified O-GlcNAc-N3 Modified Proteins and Their Functional Classificationsa identified protein

βb,c

Translation elongation factor 1 Translation elongation factor 1 γb,c eIF-4ab,c Nucleosome remodeling factorc Thioredoxin peroxidase 1 Catalasec Slow superoxide dismutase CG6776 Glutathione-S-transferase 1-1 Capulet Chd64 β tubulin 60D CG10724 Cofilin/actin depolymerizing factor Hsc70-4b Hsc70-3b Hsc70Cbb Hsp83 Proteosome alpha 6 subunit Proteosome alpha 4 subunit Prolyl oligopeptidase Puromycin sensitive aminopeptidase Ubiquitin activating enzyme 1 Protein disulfide isomerasec Rana Calreticulinc Annexin Xb Regucalcin EGFR type 3 CG10417 CG6180 14-3-3  Eb1 Pyruvate kinaseb GAPDHb Triosephosphate isomerase Myo-inositol-1-phosphate synthase CG1129 CG2767 CG6084 Ade5 CG32626 Thymidylate synthase Anon1A3 CG6259 CG5094 CG31305-PJ CG2446 CG10630 Transitional ER ATPasec CG6891

gi no.

functional classification

molecular function

13124189 6716514 418925 18447455 24641739 17981717 4572573 21355779 121694 28317136 21355917 21626738 21356527 1168731 24647034 24641402 21357475 17647529 24586400 12643270 24583414 24655260 17861718 17647799 21356159 24645441 71777 18860103 552091 19921654 24583960 24647885 19921706 3108349 84977 267156 4106368 21355443 24644950 24662781 18860083 18859679 13959716 5616364 21355253 24584835 24645977 20129011 24658349 3337433 21483504

translation translation translation transcription stress response stress response stress response stress response stress response structural structural structural structural structural chaperone, stress response chaperone, stress response chaperone, stress response chaperone. stress response protein degradation protein degradation protein degradation protein degradation protein degradation protein modification protein targeting protein transport signaling signaling signaling signaling signaling cell cycle cell cycle glucose metabolism glucose metabolism glucose metabolism carbohydrate metabolism carbohydrate metabolism carbohydrate metabolism carbohydrate metabolism nucleotide metabolism nucleotide metabolism nucleotide metabolism unknown unknown unknown unknown unknown unknown unknown unknown

Elongation factor Elongation factor Translation initiation factor Chromatin remodeling Peroxidase Peroxidase Superoxide dismutase Glutathione transferase Glutathione transferase Actin binding Actin binding Tubulin binding Actin polymerizing/depolymerizing Cofilin/actin depolymerizing Protein folding Protein folding Protein folding Protein folding Proteolysis Proteolysis Proteolysis Aminopeptidase Ubiquitin activation Disulfide isomerase Small GTPase Calcium ion binding Calcium ion binding Calcium mediated signaling EGF receptor Protein S/T phosphatase Kinase inhibitor PKC inhibitor Microtubule binding Kinase Dehydrogenase Isomerase Mannose-1-phosphate guanylyltransferase Alcohol dehydrogenase Aldehyde reductase Purine base metabolism Purine base metabolism Nucleotide synthase

a Classifications are based on NCBI and FlyBase annotations. b Proteins previously reported to be O-GlcNAc modified. c Proteins with annotated cellular localization other than the cytoplasm.

formed by the Bertozzi group on the ability for the hexosamine salvage pathway to accommodate the azide functional group.15 The azide moiety has advantageous properties for tagging a protein. For example, the azide moiety is small, inducing minimum perturbation of substrate structure. Also, the azide moiety, in the context of the GlcNAc molecule, is not toxic to the cell. Furthermore, the azide provides a tag for the selective, specific, covalent conjugation of a probe, such as biotin, via an enhanced version of the Staudinger ligation.15 By incorporating a biotin probe into the phosphine capture reagent, we were able to take advantage of the strong interaction between biotin and streptavidin. This allows the detection of azido-GlcNAc modified proteins via Western Blotting with streptavidin-conjugated HRP, and the selective isolation of azido-GlcNAc modified proteins using streptavidin-conjugated beads.

The identification of novel O-GlcNAc modified proteins will help elucidate the function for the modification. Past methods used in the study of O-GlcNAc modified proteins include the use of O-GlcNAc specific antibodies,30 wheat germ agglutinin columns,31 β-elimination coupled with Michael addition (BEMAD),32,33 or enzymatic attachment of radiolabeled or derivatized galactose molecules to O-GlcNAc modifications.1,34 Antibody-based methods using O-GlcNAc-specific monoclonal antibodies rely on a binding affinity between the antibody and its antigen. The relatively low binding affinity of antibodies against O-GlcNAc precludes the use of the harsh washing conditions. As a result, immunoaffinity purification not only isolates O-GlcNAc modified proteins but also their interaction partners. Although antibody-based methods represent a good starting point for the characterization of OGlcNAc modified proteins, the relatively low binding affinity Journal of Proteome Research • Vol. 4, No. 3, 2005 955

research articles makes antibodies an inappropriate choice for large-scale proteomic analyses. Enrichment of glycosylated proteins using wheat germ agglutinin columns is not specific for a particular sugar and the binding of wheat germ agglutinin to monosaccharide modifications is weak.31 Such methods suffer from the nonspecific binding of proteins and cannot employ the harsh washing conditions we use (8 M urea, 1% SDS/PBS) to reduce such nonspecific interactions. BEMAD involves the basecatalyzed elimination of modification groups at serine and threonine residues. The resulting R,β-unsaturated carbonyl is subsequently reacted with a nucleophilic agent such as dithiothreitol to tag the previously modified site. BEMAD has been described as a method allowing the selective isolation and site mapping of O-GlcNAc specifically,33 as well as serine/threonine posttranslational modifications in general.32 However, the harsh conditions used in BEMAD strategy can lead to degradation of peptides (data not published) and side reactions with unmodified serine and threonine residues.35 Enzymatic tagging of GlcNAc residues with radiolabeled galactose was described in the seminal paper on O-GlcNAc.1 Unfortunately, this strategy does not allow for the selective isolation of the labeled proteins for proteomic analysis. Recently, Tai et al. described a technique for the labeling of GlcNAc residues with a ketone analogue of UDP-galactose.34 The method exploits a mutant of the enzyme β-1,4-galactosyl transferase containing an active site which accommodates the ketone moiety in the substrate. The ketone allows covalent coupling of aminooxy biotin as a tag. Using this method, 29 peptides representing 25 proteins were idendified as O-GlcNAc modified in samples of rat brain, including 2 previously known proteins.36 The TAS technology has a few unique features. It can be used for both the selective detection and isolation of azido-GlcNAc modified proteins. Due to the strong interaction between biotin and streptavidin, other nonazido-GlcNAc modified proteins could be efficiently removed by harsh washing solutions such as 8 M urea and 1-2% SDS, allowing selective enrichment. Finally, the Staudinger ligation is truly bioorthogonal, since neither azides nor phosphines have been identified as constituents of any known biomolecule. Nevertheless, the TAS strategy in its current form suffers from some limitations. The current study did not address the issue of site mapping. In the interest of reducing nonspecific binding, we employed streptavidin beads. While this allows the use of harsh washing steps, it is refractory to the elution of the biotinylated peptides. Thus, peptides specifically bearing the O-GlcNAc modification were not subjected to MS identification. In addition, the current method is not a quantitative method. This could be overcome by synthesizing a heavy isotope capture reagent to complement our light isotope reagent. Avidin monomer beads would provide a selective binding strategy allowing for elution of bound peptides and quantitation of relative labeling between pooled samples from differentially labeled sources. The sites of the eluted peptides can be mapped by MS/MS/MS analysis in a similar fashion as for mapping phosphorylation sites.37 Future directions for this technology include development of a second-generation capture reagent incorporating a cleavable linker for the elution of azido-GlcNAc modified peptides from the streptavidin beads. This reagent would help identify sites of O-GlcNAc modification. Additionally, the synthesis of a heavy isotope labeled phosphine capture reagent would facilitate quantitation of modified proteins between samples. These advancements would allow a comprehensive study of 956

Journal of Proteome Research • Vol. 4, No. 3, 2005

Sprung et al.

both the dynamics of the O-GlcNAc modification and the sitespecific changes in patterns of modification that occur during the development of insulin resistance. In summary, we have demonstrated the TAS technology could be used to metabolically label, detect, isolate, and identify known and putative O-GlcNAc modified proteins. Such a metabolically based method is well suited for the analysis of changes in the pattern of modification that occur in response to a cellular stimulus or disease progression.

Acknowledgment. We would like to thank Dr. Mark Lehrman for helpful discussions through the course of this work. Y.Z. is supported by The Robert A. Welch Foundation (I-1550) and NIH (CA 107943). JF is supported by the Robert A.Welch Foundation and National Institutes of Health Grant GM 31278. Supporting Information Available: Information on the composition of the glucose-free S2 media prepared inhouse. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Torres, C. R.; Hart, G. W. Topography and polypeptide distribution of terminal N-acetylglucosamine residues on the surfaces of intact lymphocytes. Evidence for O-linked GlcNAc. J. Biol. Chem. 1984, 259, 3308-3317. (2) Zachara, N. E.; Hart, G. W. The emerging significance of O-GlcNAc in cellular regulation. Chem. Rev. 2002, 102, 431-438. (3) Zachara, N. E.; Hart, G. W. O-GlcNAc a sensor of cellular state: the role of nucleocytoplasmic glycosylation in modulating cellular function in response to nutrition and stress. Biochim. Biophys. Acta 2004, 1673, 13-28. (4) Vosseller, K.; Sakabe, K.; Wells, L.; Hart, G. W. Diverse regulation of protein function by O-GlcNAc: a nuclear and cytoplasmic carbohydrate posttranslational modification. Curr. Opin. Chem. Biol. 2002, 6, 851-857. (5) Wells, L.; Vosseller, K.; Hart, G. W. Glycosylation of nucleocytoplasmic proteins: signal transduction and O-GlcNAc. Science 2001, 291, 2376-2378. (6) Hanover, J. A. Glycan-dependent signaling: O-linked N-acetylglucosamine. Faseb J. 2001, 15, 1865-1876. (7) Haltiwanger, R. S.; Blomberg, M. A.; Hart, G. W. Glycosylation of nuclear and cytoplasmic proteins. Purification and characterization of a uridine diphospho-N-acetylglucosamine: polypeptide beta-N-acetylglucosaminyltransferase. J. Biol. Chem. 1992, 267, 9005-9013. (8) Dong, D. L.; Hart, G. W. Purification and characterization of an O-GlcNAc selective N-acetyl-beta-D-glucosaminidase from rat spleen cytosol. J. Biol. Chem. 1994, 269, 19321-19330. (9) O’Donnell, N.; Zachara, N. E.; Hart, G. W.; Marth, J. D. Ogtdependent X-chromosome-linked protein glycosylation is a requisite modification in somatic cell function and embryo viability. Mol. Cell Biol. 2004, 24, 1680-1690. (10) Wells, L.; Vosseller, K.; Hart, G. W. A role for N-acetylglucosamine as a nutrient sensor and mediator of insulin resistance. Cell Mol. Life Sci. 2003, 60, 222-228. (11) Vosseller, K.; Wells, L.; Lane, M. D.; Hart, G. W. Elevated nucleocytoplasmic glycosylation by O-GlcNAc results in insulin resistance associated with defects in Akt activation in 3T3-L1 adipocytes. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 5313-5318. (12) McClain, D. A. et al. Altered glycan-dependent signaling induces insulin resistance and hyperleptinemia. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 10695-10699. (13) Buse, M. G.; Robinson, K. A.; Marshall, B. A.; Hresko, R. C.; Mueckler, M. M. Enhanced O-GlcNAc protein modification is associated with insulin resistance in GLUT1-overexpressing muscles. Am. J. Physiol. Endocrinol. Metab. 2002, 283, E241-250. (14) Saxon, E.; Bertozzi, C. R. Cell surface engineering by a modified Staudinger reaction. Science 2000, 287, 2007-2010. (15) Vocadlo, D. J.; Hang, H. C.; Kim, E. J.; Hanover, J. A.; Bertozzi, C. R. A chemical approach for identifying O-GlcNAc-modified proteins in cells. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 91169121.

research articles

TAS for O-GlcNAc Modification (16) Prescher, J. A.; Dube, D. H.; Bertozzi, C. R. Chemical remodelling of cell surfaces in living animals. Nature 2004, 430, 873-877. (17) Kho, Y. et al. A tagging-via-substrate technology for detection and proteomics of farnesylated proteins. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 12479-12484. (18) Saxon, E. et al. Investigating cellular metabolism of synthetic azidosugars with the Staudinger ligation. J. Am. Chem. Soc. 2002, 124, 14893-14902. (19) Luchansky, S. J.; Goon, S.; Bertozzi, C. R. Expanding the diversity of unnatural cell-surface sialic acids. Chembiochem 2004, 5, 371374. (20) Saxon, E.; Bertozzi, C. R. Chemical and biological strategies for engineering cell surface glycosylation. Annu. Rev. Cell Dev. Biol. 2001, 17, 1-23. (21) Zhao, Y.; Kwon, S. W.; Anselmo, A.; Kaur, K.; White, M. A. Broad spectrum identification of cellular small ubiquitin-related modifier (SUMO) substrate proteins. J. Biol. Chem. 2004, 279, 2099921002. (22) Kwon, S. W.; Kim, S. C.; Jaunbergs, J.; Falck, J. R.; Zhao, Y. Selective enrichment of thiophosphorylated polypeptides as a tool for the analysis of protein phosphorylation. Mol. Cell Proteomics 2003, 2, 242-247. (23) Walgren, J. L.; Vincent, T. S.; Schey, K. L.; Buse, M. G. High glucose and insulin promote O-GlcNAc modification of proteins, including alpha-tubulin. Am. J. Physiol. Endocrinol. Metab. 2003, 284, E424-434. (24) Konrad, R. J.; Janowski, K. M.; Kudlow, J. E. Glucose and streptozotocin stimulate p135 O-glycosylation in pancreatic islets. Biochem. Biophys. Res. Commun. 2000, 267, 26-32. (25) Han, I.; Oh, E. S.; Kudlow, J. E. Responsiveness of the state of O-linked N-acetylglucosamine modification of nuclear pore protein p62 to the extracellular glucose concentration. Biochem. J. 2000, 350 Pt 1, 109-114. (26) Yki-Jarvinen, H.; Virkamaki, A.; Daniels, M. C.; McClain, D.; Gottschalk, W. K. Insulin and glucosamine infusions increase O-linked N-acetyl-glucosamine in skeletal muscle proteins in vivo. Metabolism 1998, 47, 449-555.

(27) Hawkins, M. et al. Role of the glucosamine pathway in fat-induced insulin resistance. J. Clin. Invest. 1997, 99, 2173-2182. (28) Inoki, K.; Zhu, T.; Guan, K. L. TSC2 mediates cellular energy response to control cell growth and survival. Cell 2003, 115, 577590. (29) Zachara, N. E. et al. Dynamic O-GlcNAc modification of nucleocytoplasmic proteins in response to stress. A survival response of mammalian cells. J. Biol. Chem. 2004, 279, 30133-30142. (30) Comer, F. I.; Vosseller, K.; Wells, L.; Accavitti, M. A.; Hart, G. W. Characterization of a mouse monoclonal antibody specific for O-linked N-acetylglucosamine. Anal. Biochem. 2001, 293, 169177. (31) Montreuil, J.; et al., in Carbohydrate Analysis: A Practical Approach; Chaplin, M. F., Kennedy, J. F., Eds.; IRL Press: New York, 1994, 181-293. (32) Vosseller, K. et al. Quantitative analysis of both protein expression and serine/threonine post-translational modifications through stable isotope labeling with dithiothreitol. Proteomics 2005, 5, 388-398. (33) Wells, L. et al. Mapping sites of O-GlcNAc modification using affinity tags for serine and threonine posttranslational modifications. Mol. Cell Proteomics 2002, 1, 791-804. (34) Tai, H. C.; Khidekel, N.; Ficarro, S. B.; Peters, E. C.; Hsieh-Wilson, L. C. Parallel identification of O-GlcNAc-modified proteins from cell lysates. J. Am. Chem. Soc. 2004, 126, 10500-10501. (35) McLachlin, D. T.; Chait, B. T. Improved beta-elimination-based affinity purification strategy for enrichment of phosphopeptides. Anal. Chem. 2003, 75, 6826-6836. (36) Khidekel, N.; Ficarro, S. B.; Peters, E. C.; Hsieh-Wilson, L. C. Exploring the O-GlcNAc proteome: direct identification of OGlcNAc-modified proteins from the brain. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 13132-13137. (37) Gruhler, A. et al. Quantitative Phosphoproteomics Applied to the Yeast Pheromone Signaling Pathway. Mol. Cell Proteomics 2005, 4, 310-327.

PR050033J

Journal of Proteome Research • Vol. 4, No. 3, 2005 957