Surface Initiated Atom Transfer Radical Polymerization: Access to

Oct 28, 2010 - Fax: (+) 86 80705155 (Y.J.Z. and X.H.Q.). ... Here, we report a new type of capillary column developed using surface initiated atom tra...
2 downloads 4 Views 767KB Size
Anal. Chem. 2010, 82, 9461–9468

Surface Initiated Atom Transfer Radical Polymerization: Access to Three Dimensional Wavelike Polymer Structure Modified Capillary Columns for Online Phosphopeptide Enrichment Weijie Qin, Wanjun Zhang, Lina Song, Yangjun Zhang,* and Xiaohong Qian* State Key Laboratory of Proteomics, Beijing Proteome Research Center, Beijing Institute of Radiation Medicine, No. 33 Life Science Park Road, Changping District, Beijing 102206, PR China Reversible phosphorylation is one of the most important post-translational modifications of proteins and a key regulator of cellular signaling pathways. Specific enrichment of phosphopeptides from proteolytic digests is a prerequisite for large scale identification of protein phosphorylation by mass spectrometry. Online enrichment of phosphopeptides attracts particular interests due to its automated operation, higher throughput and reproducibility, lower sample loss, and contamination. Here, we report a new type of capillary column developed using surface initiated atom transfer radical polymerization (SI-ATRP) for automated online phosphopeptide enrichment. SI-ATRP modification leads to a surface confined growth of three-dimensional wavelike polymer structure on the inner wall of capillary columns and, therefore, results in largely increased surface area. Furthermore, the noncross-linked flexible polymer chains grown by SI-ATRP create a large internal volume that allows phosphopeptides to penetrate into during enrichment and also facilitate the interaction between the numerous functional groups in the polymer chains and target phosphopeptides. Therefore, highly efficient and specific enrichment is achieved even for a low femtomole of phosphopeptides. The loading capacity is increased more than an order of magnitude compared with that obtained using conventional open tubular capillary columns. The SI-ATRP modified capillary column was successful applied in the online phosphoproteomics analysis of HepG2 cell lysate and resulted in 10 times improved phosphopeptide identification than the previously reported number. Finally, the SIATRP technique is compatible with a variety of functional monomers, and therefore, versatile potential applications in reverse phase, ion exchange, and affinity chromatography can be expected. As one of the most important post-translational modifications, protein phosphorylation is involved in the control of many * Corresponding author. E-mail: [email protected] (Y.J.Z.); [email protected] (X.H.Q.). Fax: (+) 86 80705155 (Y.J.Z. and X.H.Q.). 10.1021/ac1021437  2010 American Chemical Society Published on Web 10/28/2010

biological processes, such as metabolic pathways, membrane transport, and cellular signal transduction. Elucidation of the phosphorylation sites is essential for understanding the regulations of the key biological processes.1-4 Currently, mass spectrometry based methods are the most dominant techniques for large scale identification of protein phosphorylation owing to the high sensitivity, accuracy, and speed of modern mass spectrometry over conventional biochemical methods.5-9 Typically, protein mixtures extracted from cells, plasma, or tissues are proteolytically digested into peptides, because they are more amenable to mass spectrometric analysis. However, due to the low stoichiometric ratio, low ionization efficiency of phosphopeptides, and the ion suppression effect by nonphosphopeptides, isolation and enrichment of phosphopeptides from a proteolytic peptide mixture before mass spectrometric analysis is necessary; otherwise, the signals of phosphopeptides are overwhelmed by their unphosphorylated cognates.9-13 Affinity based enrichment methods using metal oxides (mainly TiO2) or immobilized metal ions (Fe3+, Zr4+, and Ti4+) are widely adopted due to their specific and efficient enrichment (1) Mann, M.; Jensen, O. N. Nat. Biotechnol. 2003, 21, 255–261. (2) Walsh, C. T.; Garneau-Tsodikova, S.; Gatto, G. J. Angew. Chem., Int. Ed. 2005, 44, 7342–7372. (3) Olsen, J. V.; Blagoev, B.; Gnad, F.; Macek, B.; Kumar, C.; Mortensen, P.; Mann, M. Cell 2006, 127, 635–648. (4) Linding, R.; Jensen, L. J.; Ostheimer, G. J.; van Vugt, M.; Jorgensen, C.; Miron, I. M.; Diella, F.; Colwill, K.; Taylor, L.; Elder, K.; Metalnikov, P.; Nguyen, V.; Pasculescu, A.; Jin, J.; Park, J. G.; Samson, L. D.; Woodgett, J. R.; Russell, R. B.; Bork, P.; Yaffe, M. B.; Pawson, T. Cell 2007, 129, 1415–1426. (5) Ficarro, S. B.; McCleland, M. L.; Stukenberg, P. T.; Burke, D. J.; Ross, M. M.; Shabanowitz, J.; Hunt, D. F.; White, F. M. Nat. Biotechnol. 2002, 20, 301–305. (6) Aebersold, R.; Mann, M. Nature 2003, 422, 198–207. (7) Beausoleil, S. A.; Villen, J.; Gerber, S. A.; Rush, J.; Gygi, S. P. Nat. Biotechnol. 2006, 24, 1285–1292. (8) Navaza, A. P.; Encinar, J. R.; Sanz-Medel, A. Angew. Chem., Int. Ed. 2007, 46, 569–571. (9) Villen, J.; Beausoleil, S. A.; Gerber, S. A.; Gygi, S. P. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 1488–1493. (10) Tao, W. A.; Wollscheid, B.; O’Brien, R.; Eng, J. K.; Li, X. J.; Bodenmiller, B.; Watts, J. D.; Hood, L.; Aebersold, R. Nat. Methods 2005, 2, 591–598. (11) Xu, X. Q.; Deng, C. H.; Gao, M. X.; Yu, W. J.; Yang, P. Y.; Zhang, X. M. Adv. Mater. 2006, 18, 3289–3293. (12) Li, S. W.; Zeng, D. X. Angew. Chem., Int. Ed. 2007, 46, 4751–4753. (13) Sturm, M.; Leitner, A.; Smatt, J. H.; Linden, M.; Lindner, W. Adv. Funct. Mater. 2008, 18, 2381–2389.

Analytical Chemistry, Vol. 82, No. 22, November 15, 2010

9461

of phosphopeptides. However, most enrichment methods are performed manually using packed microcolumns or on-plate enrichment, which are inherently labor intense and may results in irreproducible results.14-27 For proteome-wide phosphorylation profiling, automated online enrichment and identification of phosphopeptides is highly desired, due to its high throughput and reproducibility, lower sample loss, and contamination.28 Though online phosphopeptide enrichment using a packed capillary column or open tubular capillary column has been reported,28,29 a few inherent problems limited their application in proteomics studies. First, frits are needed in packed columns to sustain the chromatographic materials in the columns and, therefore, complicate the preparation procedure. Second, the packed beads may break after repeated usage and result in poor reproducibility and shorter usage life. Third, the relatively high back pressure of packed columns and also leads to a higher requirement to the HPLC system in online application. For open tabular columns, no successful application in complex proteomics samples has been reported due to their low enrichment efficiency and low loading capacity. Surface initiated atom transfer radical polymerization (SIATRP) is one of the most widely employed “grafting from” strategies30 andhasbeenappliedinthesynthesisofnanoparticles,31-33 surface patterning,34-39 and electrosmotic flow manipulation (14) Larsen, M. R.; Thingholm, T. E.; Jensen, O. N.; Roepstorff, P.; Jorgensen, T. J. D. Mol. Cell. Proteomics 2005, 4, 873–886. (15) Zhou, H. J.; Xu, S. Y.; Ye, M. L.; Feng, S.; Pan, C.; Jiang, X. G.; Li, X.; Han, G. H.; Fu, Y.; Zou, H. F. J. Proteome Res. 2006, 5, 2431–2437. (16) Feng, S.; Ye, M. L.; Zhou, H. J.; Jiang, X. G.; Jiang, X. N.; Zou, H. F.; Gong, B. L. Mol. Cell. Proteomics 2007, 6, 1656–1665. (17) Xu, S. Y.; Whitin, J. C.; Yu, T. T. S.; Zhou, H. J.; Sun, D. Z.; Sue, H. J.; Zou, H. F.; Cohen, H. J.; Zare, R. H. Anal. Chem. 2008, 80, 5542–5549. (18) Thingholm, T. E.; Jensen, O. N.; Robinson, P. J.; Larsen, M. R. Mol. Cell. Proteomics 2008, 7, 661–671. (19) Ficarro, S. B.; Parikh, J. R.; Blank, N. C.; Marto, J. A. Anal. Chem. 2008, 80, 4606–4613. (20) Chang, C. K.; Wu, C. C.; Wang, Y. S.; Chang, H. C. Anal. Chem. 2008, 80, 3791–3797. (21) Bi, H. Y.; Qiao, L.; Busnel, J. M.; Devaud, V.; Liu, B. H.; Girault, H. H. Anal. Chem. 2009, 81, 1177–1183. (22) Hu, L. H.; Zhou, H. J.; Li, Y. H.; Sun, S. T.; Guo, L. H.; Ye, M. L.; Tian, X. F.; Gu, J. R.; Yang, S. L.; Zou, H. F. Anal. Chem. 2009, 81, 94–104. (23) Ficarro, S. B.; Adelmant, G.; Tomar, M. N.; Zhang, Y.; Cheng, V. J.; Marto, J. A. Anal. Chem. 2009, 81, 4566–4575. (24) Hoang, T.; Roth, U.; Kowalewski, K.; Belisle, C.; Steinert, K.; Karas, M. Anal. Chem. 2010, 82, 219–228. (25) Niklew, M. L.; Hochkirch, U.; Melikyan, A.; Moritz, T.; Kurzawski, S.; Schluter, H.; Ebner, I.; Linscheid, M. W. Anal. Chem. 2010, 82, 1047– 1053. (26) Eriksson, A.; Bergquist, J.; Edwards, K.; Hagfeldt, A.; Malmstrom, D.; Hernandez, V. A. Anal. Chem. 2010, 82, 4577–4583. (27) Leitner, A.; Sturm, M.; Hudecz, O.; Mazanek, M.; Smatt, J. H.; Linden, M.; Lindner, W.; Mechtler, K. Anal. Chem. 2010, 82, 2726–2733. (28) Pinkse, M. W. H.; Mohammed, S.; Gouw, L. W.; van Breukelen, B.; Vos, H. R.; Heck, A. J. R. J. Proteome Res. 2008, 7, 687–697. (29) Xue, Y. F.; Wei, J. Y.; Han, H. H.; Zhao, L. Y.; Cao, D.; Wang, J. L.; Yang, X. M.; Zhang, Y. J.; Qian, X. H. J. Chromatogr., B-Anal. Technol. Biomed. Life Sci. 2009, 877, 757–764. (30) Matyjaszewski, K.; Tsarevsky, N. V. Nature Chem. 2009, 1, 276–288. (31) He, T.; Adams, D. J.; Butler, M. F.; Yeoh, C. T.; Cooper, A. I.; Rannard, S. P. Angew. Chem., Int. Ed. 2007, 46, 9243–9247. (32) Li, D. X.; Cui, Y.; Wang, K. W.; He, Q.; Yan, X. H.; Li, J. B. Adv. Funct. Mater. 2007, 17, 3134–3140. (33) Liu, B.; Wei, W.; Qu, X. Z.; Yang, Z. H. Angew. Chem., Int. Ed. 2008, 47, 3973–3975. (34) Ma, H. W.; Wells, M.; Beebe, T. P.; Chilkoti, A. Adv. Funct. Mater. 2006, 16, 640–648. (35) Wu, Y. Z.; Huang, Y. Y.; Ma, H. W. J. Am. Chem. Soc. 2007, 129, 7226– 7227.

9462

Analytical Chemistry, Vol. 82, No. 22, November 15, 2010

in capillary electrophoresis.40,41 Compared with the “grafting to” method, in which long polymer chains are directly attached on the substrate, the SI-ATRP “grafting from” technique gives significantly increased surface grafting density and well controlled polymer structure and thickness, since the grafting is highly controllable by in situ growth of polymer chains from the initiator immobilized surface via living/controlled polymerization. In this work, we developed a new type of capillary column using the SI-ATRP technique and successfully applied it in automated online phosphopeptide enrichment. SI-ATRP modification leads to surface confined growth of threedimensional (3D) wavelike polymer structure on the inner wall of capillary columns and, therefore, results in a largely increased surface area for target capturing. The SI-ATRP modified capillary columns are subsequently derivatized with zirconium phosphonate (ZrPO3) and coupled to LC-MS systems for automated online phosphopeptide enrichment and identification. Two advantages are obtained using this SI-ATRP technique in capillary column modification. First, since no cross-linkage is applied to linear polymer chains grown by SI-ATRP, a large internal volume between the polymer chains is created to allow target phosphopeptides to penetrate into and, thus, extends the enrichment to a three-dimensional state. Second, the noncross-linked flexible polymer chains facilitate the interaction between the numerous ZrPO3 groups in the polymer chains and target phosphopeptides. As a consequence, improved enrichment efficiency and loading capacity is achieved with relatively low column back pressure and extended usage life. EXPERIMENTAL SECTION Materials and Reagents. Bovine R-casein, bovine serum albumin (BSA), glycidyl methacrylate (GMA), 1,2-ethylenediamine, copper(II) chloride, cuprous chloride, trifluoroacetic acid (TFA), 2,5-dihydroxybenzoic acid (2,5-DHB), and phosphorus oxychloride (POCl3) were obtained from Sigma (St. Louis, MO, USA). The standard phosphopeptide (FL[pT]EYVATR, m/z ) 1179.55) was synthesized by SBS Genetech Co. Ltd. (Beijing, China). Sequencing grade porcine trypsin was purchased from Promega (Madison, WI, USA). Fused silica capillaries (i.d. 75 µm) were obtained from Yongnian (Hebei, China). Zirconyl chloride octahydrate (ZrOCl2 · 8H2O), 3-aminopropyl-triethoxysilane, R-bromoisobutyryl bromide, N,N,N′,N′′,N′′-pentamethyldiethylenetriamine, and 2,4,6-collidine were purchased from Acros. Deionized water (with resistance >18MΩ/cm) was prepared using a Millipore purification system (Billerica, MA, USA) and used throughout this work. (36) Wang, X. J.; Bohn, P. W. Adv. Mater. 2007, 19, 515–520. (37) Raynor, J. E.; Petrie, T. A.; Garcia, A. J.; Collard, D. M. Adv. Mater. 2007, 19, 1724–1728. (38) Hucknall, A.; Kim, D. H.; Rangarajan, S.; Hill, R. T.; Reichert, W. M.; Chilkoti, A. Adv. Mater. 2009, 21, 1968–1971. (39) Chen, T.; Zhong, J. M.; Chang, D. P.; Carcia, A.; Zauscher, S. Adv. Mater. 2009, 21, 1825–1829. (40) Huang, X. Y.; Doneski, L. J.; Wirth, M. J. Anal. Chem. 1998, 70, 4023– 4029. (41) Feldmann, A.; Claussnitzer, U.; Otto, M. J. Chromatogr., B-Anal. Technol. Biomed. Life Sci. 2004, 803, 149–157.

Preparation of SI-ATRP Modified and Zirconium Phosphonate Derivatized (SI-ATRP-ZrPO3) Capillary Column. First, the initiator, 3-(2-bromoisobutyramido)propyl(triethoxy)silane (BIBAPTES) carrying a triethoxysilane group was synthesized according to the previously reported procedure.42 Next, the capillary columns were treated with BIBAPTES for initiator immobilization on the inner wall of the columns. After 10 h, excess initiators were removed by repeated washing with methanol and dried by N2. For SI-ATRP grafting on the inner wall of the capillary columns, a stock solution of 2 M GMA dissolved in cyclohexanol was added in a Schlenk flask and was degassed via three freeze-pump-thaw cycles. Next, 0.01 M CuCl, 0.001 M CuCl2, and 0.015 M N,N,N′,N′′,N′′-pentamethyldiethylenetriamine were added to the flask. The flask was then purged with nitrogen for 10 min to remove oxygen and sonicated for 5 min to yield a homogeneous mixture. The SI-ATRP grafting was carried out by injecting the stock solution into the BIBAPTES treated columns, sealed, and allowed to stand at RT for times ranging from 12 to 20 h. After polymerization, the capillary column was washed with dimethyl sulfoxide to remove excess reagents, dried with N2, and then treated with 1,2-ethylenediamine at 80 °C for 4 h for amination via the well-known ring-opening reaction of an epoxy group by an amine group.43,44 The aminated capillary column was further derivatized by 0.2 M POCl3 and 0.2 M 2,4,6-collidinein in anhydrous acetonitrile for 0.5 h to yield phosphonate. Next, the phosphorylated column was treated with 0.1 M ZrClO2 aqueous solution for 0.5 h for zirconium deposition.45,46 The capillary column was washed with acetonitrile and dried with N2 after each treatment. Finally, the ZrPO3 derived capillary column was rinsed with deionized water, sealed, and stored at 4 °C for future use. Capillary columns modified using conventional free radical polymerization was prepared as a control. Briefly, γ-methacryloxypropyltrimethoxysilane (γ-MPS) was immobilized on the inner wall of capillary columns. After removal of the excess reagent, 2 M GMA containing 1% azobisisobutyronitrile (w/w) was injected into the columns. The columns were sealed and placed at 70 °C for 16 h for polymerization. Tryptic Digestion of Standard Proteins. Bovine R-casein (500 µg) was digested in 500 µL of 50 mM ammonium bicarbonate buffer (pH 8.0) using trypsin at a substrate/trypsin ratio of 50:1 and incubated for 16 h at 37 °C. BSA (500 µg) was dissolved in 500 µL of 50 mM ammonium bicarbonate containing 8 M urea. Then, dithiothreitol (DTT) and indole-3-acetic acid (IAA) were sequentially added into the BSA solution. After eight times dilution of the BSA solution with 50 mM ammonium bicarbonate, trypsin was introduced into the solution at a substrate/trypsin ratio of 50:1 and incubated at 37 °C for 16 h. Thirty microliters of bovine milk was diluted with 900 µL of 50 mM ammonium bicarbonate, and the solution was centrifuged at 16 000 rpm for 30 min. The supernatant was concentrated using SpeedVac (Thermo Savant), (42) Tugulu, S.; Arnold, A.; Sielaff, I.; Johnsson, K.; Klok, H. A. Biomacromolecules 2005, 6, 1602–1607. (43) Allmer, K.; Hult, A.; Ranby, B. J. Polym. Sci., Part A: Polym. Chem. 1989, 27, 1641–1652. (44) Kubota, H.; Ujita, S. J. Appl. Polym. Sci. 1995, 56, 25–31. (45) Horne, J. C.; Blanchard, G. J. J. Am. Chem. Soc. 1999, 121, 4427–4432. (46) Horne, J. C.; Huang, Y.; Liu, G. Y.; Blanchard, G. J. J. Am. Chem. Soc. 1999, 121, 4419–4426.

followed by denaturation in 8 M urea solution, reduction with DTT for 4 h at 37 °C, and alkylation with IAA for 60 min at room temperature in the dark. Finally, trypsin was added at a mass ratio of 1:50 (enzyme/protein), and the mixture was incubated at 37 °C for 16 h. All the tryptic digests were stored at -20 °C for future use. Automated Enrichment and Identification of Phosphopeptides Using LC-MALDI. The SI-ATRP-ZrPO3 column was integrated into a high performance liquid chromatography system (LC Packings/Dionex, CA, USA) for automated phosphopeptide enrichment. The tryptic digested peptides in 80% acetonitrile (ACN) solution containing 0.1% TFA was infused into the SI-ATRP-ZrPO3 capillary column at a flow rate of 1 µL/min. After that, 50 µL of 80% ACN solution containing 0.1% TFA was injected into the SI-ATRP-ZrPO3 column at a flow rate of 1 µL/ min to remove the nonphosphopeptides which were nonspecifically bound to ZrPO3. Finally, ammonia solution (0.1 M, pH ) 11) was infused into the SI-ATRP-ZrPO3 column to elute the captured phosphopeptides. The eluate containing phosphopeptides was continuously mixed with 10 mg/mL DHB dissolved in 50% acetonitrile and 1% phosphoric acid and spotted onto a MALDI target using a robot system (Probot Microfraction Collector, LCPackings/Dionex) for MALDI-TOF MS analysis. Determination of the Loading Capacity of SI-ATRP-ZrPO3 Columns on Phosphopeptides. The loading capacity of SIATRP-ZrPO3 columns was determined using either the total ion current plot of synthetic phosphopeptide (FL[pT]EYVATR) in mass spectrometry or frontal chromatography analysis. The detailed experimental procedures are shown in Supporting Information (Figure S-1). HepG2 Cell Culture and Lysate. HepG2 cells were grown at 37 °C in Dulbecco’s modified eagle medium (DMEM) supplemented with 10% v/v heat-inactivated fetal calf serum, 25 U/mL penicillin, and 25 µg/mL streptomycin. The harvested cells were washed twice with ice cold phosphate buffered saline (PBS), resuspended in ice cold lysis buffer containing 50 mM ammonium bicarbonate (pH 8.2), 8 M urea, and Sigma-Aldrich phosphatase inhibitor cocktails I and II and sonicated. Cell debris was removed by centrifugation at 12000g for 15 min at 10 °C. The supernatants were collected, and the protein concentration was determined with the Bradford assay (Bio-Rad). Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis Fractionation and Trypsin Digestion. To reduce sample complexity, the total cell lysis protein of HepG2 cells was fractioned using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) before online enrichment of phosphopetides. Briefly, 400 µg of protein was separated by 4-10% SDS-PAGE gels and stained with Coomassie Bright Blue G-250. The gel was manually cut into 10 slices. Next, each gel slice was cut into approximately 1 mm3 pieces and destained in 50% ACN and 25 mM ammonium bicarbonate. The gel pieces were dehydrated in 100% ACN and dried under vacuum. Ingel digestion was performed by washing the gel pieces with water and rehydrating with 20 mM ammonium bicarbonate (pH 7.8), reduced with 10 mM DTT, alkylated with 55 mM iodoacetamide, and digested with trypsin overnight at 37 °C. The tryptic digested peptides were extracted from gel pieces with 5% TFA and 2.5% TFA in 50% ACN twice at 37 °C. The Analytical Chemistry, Vol. 82, No. 22, November 15, 2010

9463

peptides extracts were dried under vacuum and stored at -80 °C. Online Enrichment and Identification of Phosphopeptides Using NanoLC-ESI-MS/MS. The SI-ATRP-ZrPO3 capillary column was integrated into an Eksigent nanoLC 2D system coupled with a hybrid linear ion trap-7 T Fourier transform ion cyclotron resonance mass spectrometer (LTQ-FTMS) for online phosphopeptide enrichment and identification. The schematic overview of the vented column switching setup is shown in Figure S-2 (Supporting Information). When the ten-port valve is in position one (connected between 1 and 2), the solution loaded by the autosampler flows through the SIATRP-ZrPO3 column and goes into a waste vial, while the solution goes to the PepMap C18 precolumn (320 µm × 5 mm, LC Packings, Amsterdam, The Netherlands), when the ten-port valve is in position two (connected between 1 and 10). The peptide mixtures obtained from SDS-PAGE fractionation and in gel digestion were loaded on the SI-ATRPZrPO3 column in 80% ACN containing 0.1% TFA by the autosampler and washed with 80% ACN containing 0.1% TFA at a flow rate of 1 µL/min. Phosphopeptides were trapped in a SI-ATRP-ZrPO3 column, while the nonphosphopeptides were flowed through and trapped in the C18 precolumn, when the ten-port valve was in position one. After switching the valve to position two, the nonphosphopeptides in the C18 precolumn were eluted to an analytical C18 column (75 µm × 150 mm, LC Packings) for MS analysis. Next, the phosphopeptides trapped in the SI-ATRP-ZrPO3 column were eluted to the C18 precolumn by injection of 20 µL of 100 mM NH3 · H2O (pH ) 11), desalted, and eluted onto an analytical C18 column. The chromatography separation used a gradient from solution A, (2% ACN/98% water) to solution B (80% ACN/20% water), both containing 0.1% formic acid, over 136 min at a flow rate of 300 nL/min. Characterization. The cross section structure and internal morphology of the capillary columns was characterized by a Hitachi S-4800 cold field emission scanning electron microscopy (FESEM). MALDI-TOF-MS experiments were performed on a 4800 MALDI-TOF-TOF analyzer (Applied Biosystems, MA, USA) equipped with a Nd:YAG laser at excitation wavelength 355 nm. All the mass spectra (1000 laser shots for every spectrum) were obtained in positive reflection mode after being calibrated using an external calibration with known m/z values of peptides originated from myoglobin and analyzed by Data Explorer (Version 4.5). The nanoLC-MS/MS analysis was carried out on an Eksigent nanoLC 2D system coupled with a hybrid linear ion trap-7 T Fourier transform ion cyclotron resonance mass spectrometer, LTQ-FTMS (Thermo-Electron). The spray voltage was set at 1.8 kV. All MS and MS2 spectra were acquired in data-dependent mode, and the mass spectrometer was set as a full scan MS followed by 10 data-dependent MS/MS scans. Database Searching. The acquired raw files from LTQ-FTMS were converted to .dta files by BioWorks 3.2. All the .dta files were merged into one .mgf file by a script named merge .pl. (Matrix Science Ltd.). The processed files were subsequently searched against the human sequence library in the International Protein Index (IPI) protein sequence database (IPI human database version 3.53) using an in-house Mascot server (version 2.2.0; Matrix Science Ltd., 9464

Analytical Chemistry, Vol. 82, No. 22, November 15, 2010

Scheme 1. Preparation of the SI-ATRP Modified and ZrPO3 Derivatized Enrichment Capillary Column

London, UK). Trypsin was chosen as the proteolytic enzyme, and up to two miss cleavages were allowed. Carbamidomethyl (Cys) was set as the fixed modification. Oxidation (Met) and phosphorylation (STY) were set as the variable modifications. The mass tolerance of the precursor ion was set to 20 ppm and that of the fragment ions was set to 0.8 Da. For Mascot searches, peptides with p < 0.05 and score g35 were selected.

RESULTS AND DISCUSSION Preparation and Structure Characterization of the SIATRP Modified Capillary Column. The preparation procedure of the SI-ATRP modified and zirconium phosphonate derivatized capillary column (SI-ATRP-ZrPO3) is illustrated in Scheme 1. Briefly, 3-(2-bromoisobutyramido)propyl(triethoxy)-silane (the initiator for SI-ATRP) carrying a triethoxysilane group is immobilized on the inner wall of capillary columns via the Si-O bond. Next, SI-ATRP is carried out by injecting glycidyl methacrylate (GMA) with CuCl, CuCl2, and N,N,N′,N′′,N′′pentamethyldiethylenetriamine on the column. Noncrosslinked linear polymer chains are grown from the inner wall of capillary columns by SI-ATRP grafting. After reaching the preset grafting time, the epoxy groups on the obtained GMA polymer are converted to ZrPO3 by sequential treatment with 1,2-ethylenediamine, POCl3, and ZrClO2. The inner wall of the obtained SI-ATRP-ZrPO3 columns (12 and 20 h SI-ATRP grafting) were characterized by scanning electron microscopy (SEM) and compared with a bare capillary column without SI-ATRP grafting (Figure 1a,b,c). Figure 1a is the cross section view of a SI-ATRP-ZrPO3 column with 12 h SIATRP grafting. 3D wavelike polymer structure is grown on the entire inner wall of the column. The polymer structure is solely grown from the inner wall with no bulk structure formation. The thickness of the polymer structure increases from approximately 2 to 3 µm (Figure 1a) to 3-5 µm (Figure 1b) with increasing SI-ATRP grafting time from 12 to 20 h. We attribute this roughly proportional increase of the thickness of polymer structure with increasing grafting time to the controlled polymer growth by SI-ATRP, in which polymer chain transfer and termination is minimized. Another key advantage of SI-ATRP is the ability of conducting surface confined growth of polymer structure from the inner wall of capillary columns, which is hard to achieve using conventional free radical polymerization. In conventional free radical polymerization, polymer growth is not restricted to the inner wall (Figure S-3, Supporting Information), since the initiators are dissolved in the polymerization solution instead of being immobilized on

Figure 1. SEM characterization of SI-ATRP-ZrPO3 columns with 12 h (a) and 20 h of grafting (b) and a bare column without modification (c).

the inner wall. Comparison on the surface morphology of SIATRP-ZrPO3 columns and a bare one without grafting reveals obviously increased surface roughness in SI-ATRP-ZrPO3 columns, as wavelike surface with lots of irregular nanometer sized globules is found throughout their inner wall. This type of structure is highly beneficial for target enrichment, since it not only largely increases the surface area for target binding but also reduces the steric hindrance in surface reaction.

Phosphopeptide Enrichment Using SI-ATRP-ZrPO3 Column and Enrichment Efficiency, Reproducibility, Specificity, and Loading Capacity Determination. The SI-ATRPZrPO3 column was integrated into a high performance liquid chromatography system equipped with a Probot Microfraction Collector for automated phosphopeptide enrichment and spotting. Tryptic digest of standard phosphorylated protein of R-casein was used as the model sample to evaluate the ability of SI-ATRP-ZrPO3 columns for phosphopeptide enrichment. MALDI mass spectra of tryptic digest of 5 pmol of R-casein from direct analysis and after SI-ATRP-ZrPO3 column enrichment are shown in Figure 2a,b. MALDI analysis without enrichment results in many intensive nonphosphopeptide peaks (Figure 2a). Only four weak phosphopeptide peaks are identified in Figure 2a, due to the strong ionization suppression from the nonphosphopeptides. In contrast, significantly improved phosphopeptide identification is obtained with the same amount of sample after SI-ATRP-ZrPO3 column enrichment (Figure 2b). Signals from nonphosphopeptides are almost completely removed. Totally, 22 phosphopeptides including 9 mono- and 13 multiphosphopeptides are identified with obviously enhanced signal intensity. The identified phosphopeptides cover all the theoretical phosphorylation sites in R-casein, indicating that highly efficient and unbiased enrichment is achieved. The sequences of the identified phosphopeptides labeled in Figure 2b are given in Table 1. Decreasing the loading amount of tryptic digest of R-casein to 50 and 5 fmol still results in the identification of 17 and 12 phosphopeptides (Figure 2c,d). We noticed that the S/N ratio of phosphopeptides carrying 4 to 5 phosphate groups (Nos. 15, 16, 18, and 20 in Figure 2d) are relatively low for 5 fmol of loading amount. This is probably due to the high hydrophilic property of multiphosphopeptides, which leads to particularly low ionization efficiency in MS. However, considering the lowfemtomole loading amount of peptide mixture, this enrichment efficiency is quite high. We attribute this highly efficient enrichment to the SI-ATRP induced 3D wavelike polymer structure on the inner surface of SI-ATRP-ZrPO3 columns, which leads to a largely increased surface area for target binding. Furthermore, since no cross-link reagent is used, the linear polymer chains grown by SI-ATRP has high chain flexibility. Combining these two merits with the high density of functional ZrPO3 in the polymer chains, much higher collision opportunities between ZrPO3 and phosphopeptides are obtained. This exceptionally high target capturing efficiency of the SI-ATRP-ZrPO3 column can be further demonstrated by comparison with phosphopeptide enrichment using a monolayer of Zr4+ modified open tubular capillary columns, in which 40 pmol of R-casein is needed for the identification of 16 phosphopeptides.29 The reproducibility of SI-ATRP-ZrPO3 column enrichment was evaluated by 15 repeated enrichment runs with 5 pmol of tryptic digest of R-casein. The numbers of phosphopeptides identified in each run are shown in Figure S-4, Supporting Information. The relative standard deviation (RSD) of the number of identified phosphopeptides in 15 enrichment runs is 5.69%. The number of identified phosphopeptides starts to drop after every four or five runs, which is attributed to the Analytical Chemistry, Vol. 82, No. 22, November 15, 2010

9465

Figure 2. MALDI mass spectra of tryptic digest of 5 pmol of R-casein from direct analysis (a) and enriched by the SI-ATRP-ZrPO3 column (b) and of tryptic digest of 50 fmol of R-casein enriched by SI-ATRP-ZrPO3 column (c) and of tryptic digest of 5 fmol of R-casein enriched by SI-ATRP-ZrPO3 column (d).

zirconium loss in the elution step. Therefore, a renew step using 0.1 M ZrClO2 is performed to resume the capturing capability of SI-ATRP-ZrPO3 columns after every five runs. With this treatment, enrichment of 22 phosphopeptides can still be reached after 100 repeated enrichment runs using the same column (data not shown). 9466

Analytical Chemistry, Vol. 82, No. 22, November 15, 2010

The ability to specifically enrich phosphopeptides in the presence of a huge amount of nonphosphopeptides is a key issue for phosphoproteome analysis. To further evaluate the enrichment specificity of SI-ATRP-ZrPO3 columns, a more complex sample mixture with a substantial fraction of nonphosphopeptides was used. Tryptic digest of R-casein and BSA (a nonphosphorylated

Table 1. Identified Phosphopeptides of the Tryptic Digest of r-Casein Enriched by the SI-ATRP-ZrPO3 Columna

a

no.

peptide sequence

number of phosphate groups

(M + H)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

EQL[pS]T[pS]EENSK S2-(141-151) EQL[pS]T[pS]EENSK S2-(141-151) TVDME[pS]TEVFTK S2-(153-164) TVD[Mo]E[pS]TEVFTK S2-(153-164) EQL[pS]T[pS]EENSKK S2-(141-152) TVDME[pS]TEVFTKK S2-(153-165) VPQLEIVPN[pS]AEER S1-(121-134) YLGEYLIVPN[pS]AEER S1-(104-119) DIG[pS]E[pS]TEDQAMEDIK S1-(58-73) DIG[pS]E[pS]TEDQAMEDIK S1-(58-73) DIG[pS]E[pS]TEDQA[Mo]EDIK S1-(58-73) YKVPQLEIVPN[pS]AEER S1-(119-134) NMAINPSpKENLCSTFCK S2-(40-56) NTMEHV[pS] [pS] [pS]EE[pS]IISQETYK S2-(17-36) VNEL[pS]KDIG[pS]E[pS]TEDQAMEDIK S1-(52-73) Q*MEAE[pS]I[pS][pS][pS]EEIVPN[pS]VEAQK S1-(74-94) QMEAE[pS]I[pS] [pS] [pS]EEIVPN[pS]VEAQK S1-(74-94) NTMEHV[pS] [pS] [pS]EE[pS]IISQETYKQ S2-(17-37) EKVNEL[pS]KDIG[pS]E[pS]TEDQAMEDIK S1-(50-73) NANEEEYSIG[pS] [pS] [pS]EE[pS]AEVATEEVK S2-(61-85) NANEEEY[pS]IG[pS][pS][pS]EE[pS]AEVATEEVK S2-(61-85) KNTMEHV[pS] [pS] [pS]EE[pS]IISQETYKQEK S2-(16-39)

1 2 1 1 2 1 1 1 1 2 2 1 1 4 3 5 5 4 3 4 5 4

1331.62 1411.59 1466.72 1482.7 1539.7 1594.82 1660.92 1832.83 1847.86 1927.84 1943.81 1952.12 1979.44 2619.15 2678.24 2704.25 2721.16 2747.23 2935.42 3008.28 3089.27 3132.2

[pS]: phosphorylated serine; [Mo]: oxidation on methionine; Q*: pyroglutamylation on the N-terminal Gln.

protein) at ratio of 1:100 (R-casein: BSA) was prepared and enriched by the SI-ATRP-ZrPO3 column. MALDI mass spectra obtained from direct analysis and after enrichment are displayed in Figure 3. Without enrichment, the signals of phosphopeptides are completely overwhelmed by the highly abundant nonphosphopeptide peaks and no phosphopeptides can be identified (Figure 3a). While, for the enriched sample, 19 phosphopeptides are easily detected with a clear background (Figure 3b) demonstrating that the high enrichment specificity of SI-ATRPZrPO3 columns is not impaired by the presence of highly abundant nonphosphopeptides. We noticed that a few nonphosphopeptide peaks are shown in Figure 3b, such as 1502.61, 1554.65, 1749.66, and 2458.18. These interfering nonphosphopeptides all contain three to six acidic amino acids such as aspartic acid or glutamic acid in their sequences (Table S-1, Supporting Information). The carboxyl groups on the side chain of these two types of amino acids may have weak coordination with ZrPO3.29 However, we think the enrichment specificity is still quite high considering the concentration of phosphopeptides is 2 orders of magnitude lower than that of the nonphosphopeptides. The high enrichment specificity is mainly attributed to two possible reasons. First, the high coordination affinity of ZrPO3 to phosphonate results in highly specific capturing of phosphopeptides.15,17,24 Second, compared with two-dimensional surface capturing in open tubular capillary columns, the noncross-linked three-dimensional polymer structure in SI-ATRP-ZrPO3 columns offers significantly increased target binding sites; therefore, phosphopeptides do not have to compete with acidic nonphosphopeptides for binding sites and are lost during enrichment. This is especially the case for single phosphopeptides which have relatively weaker binding affinity. The loading capacity of SI-ATRP-ZrPO3 columns is determined using either total ion current plot of synthetic phosphopeptide (FL[pT]EYVATR) in mass spectrometry or frontal chromatography analysis. The loading capacities obtained from the two tests are 131.5 and 114.4 pmol (75 µm i.d. × 50 cm),

respectively. Both of them are more than 1 order of magnitude higher than that obtained in a conventional open tubular capillary column.29 This largely increased loading capacity is attributed to the large internal volume between the noncrosslinked linear polymer chains on the inner surface of SI-ATRPZrPO3 columns that allows target phosphopeptides to penetrate into during enrichment. Though the loading capacity of the SI-ATRP column is still lower than that of the monolithic column reported by Hou C. Y. et al,47 since this work is focused on phosphopeptides, which are generally in very low abundance in real samples, a 10 times increase of loading capacity of the SI-ATRP column compared with the conventional open tubular capillary column should satisfy the need for online enrichment. Application of SI-ATRP-ZrPO3 Columns in the Online Enrichment and Identification of Phosphopeptides of HepG2 Cell Lysate. To evaluate the online phosphopeptide capturing performance of the SI-ATRP-ZrPO3 column with complex biological samples, the column was coupled to a nanoLCESI-MS/MS system and applied in the phosphoproteome analysis of HepG2 total cell lysate. The schematic overview of the vented column switching setup is shown in Figure S-2 (Supporting Information). A recovery step after every five enrichment runs was included by injection of 0.1 M ZrClO2 to redeposit zirconium in the column. In this way, the SI-ATRPZrPO3 column is completely regenerated and allows automated multiple enrichment runs. HepG2 cell lysate (400 µg) was subjected to SDS-PAGE prefractionation to reduce the sample complexity. After in gel digestion and extraction, the fractionated peptide mixtures were then enriched using the SI-ATRPZrPO3 column and analyzed by an LTQ-FT mass spectrometer in an automated online mode. A typical chromatogram of (47) Hou, C. Y.; Ma, J. F.; Tao, D. Y.; Shan, Y. C.; Liang, Z.; Zhang, L. H.; Zhang, Y. K. J. Proteome Res. , 9, 4093–4101. (48) Lee, H. J.; Na, K.; Kwon, M. S.; Kim, H.; Kim, K. S.; Paik, Y. K. Proteomics 2009, 9, 3395–3408. (49) Paradela, A.; Albar, J. P. J. Proteome Res. 2008, 7, 1809–1818.

Analytical Chemistry, Vol. 82, No. 22, November 15, 2010

9467

Figure 3. MALDI mass spectra of tryptic digest mixtures of R-casein and BSA (1:100) without (a) and with (b) enrichment with the SI-ATRPZrPO3 column.

HepG2 cell extracts is shown in the Supporting Information (Figure S-5). Unique phosphopeptides (475) containing 662 phosphorylation sites were identified (Table S-2, Supporting Information). This number is obviously higher than previously reported using online TiO2 based enrichment, in which only 39 unique phosphopeptides were identified in HepG2 cell lysate;48 therefore, largely improved phosphopeptide enrichment is obtained using the SI-ATRP-ZrPO3 column. Among the identified phosphorylation sites, 89.1% were located at serine residues, 10.1% at threonine residues, and 0.8% at tyrosine residues, which are consistent with those reported in the literature49 indicating there is no bias in the enrichment. CONCLUSION In conclusion, a new type of capillary column modified with 3D wavelike polymer structure has been successfully developed using the SI-ATRP technique. Compared with existing online enrichment methods, our method has the combined advantages of well controlled capillary column modification, high loading capacity, and unbiased enrichment of an extremely low amount of phosphopeptides. Finally, we want to emphasize the versatile potential applications of the SI-ATRP technique for capillary column modification in reverse phase, ion exchange, and affinity

9468

Analytical Chemistry, Vol. 82, No. 22, November 15, 2010

chromatography, since SI-ATRP is compatible with a variety of functionalized monomers and does not require stringent experimental conditions.30 ACKNOWLEDGMENT W.J.Q. and W.J.Z. contributed equally to this work. This work was supported by the National Natural Science Foundation of China (Grant Nos. 20635010, 20735005, 20875101, and 20905077), National Key Program for Basic Research Grants (Nos. 2006CB910801, 2007CB914104, 2010CB912701, and 2011CB910600), the National High-Tech Research and Development Program (Nos. 2006AA02A308 and 2008AA02Z309), the State Key Laboratory of Proteomics (SKLPK200807 and SKLP-0200808), and the National Key Technologies R&D Program for New Drugs (2009ZX09301-002). SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

Received for review August 14, 2010. Accepted October 15, 2010. AC1021437