Divinylbenzene)-IDA-FeIII in

The main key to succes in phosphoproteome analysis is the selective enrichment of phosphopeptides in the presence of a large number of nonphosphorylat...
17 downloads 0 Views 600KB Size
Download PDF
Poly(Glycidyl Methacrylate/Divinylbenzene)-IDA-FeIII in Phosphoproteomics Nurul H. Aprilita,†,§ Christian W. Huck,*,† Rania Bakry,† Isabel Feuerstein,† Guenther Stecher,† Sandra Morandell,‡ Hong-Lei Huang,‡ Taras Stasyk,‡ Lukas A. Huber,‡ and Guenther K. Bonn† Institute of Analytical Chemistry and Radiochemistry, Leopold-Franzens University, Innrain 52a, 6020-Innsbruck, Austria, and Biocenter, Division of Cell Biology, Innsbruck Medical University, Mu ¨ llerstraβe 59, 6020-Innsbruck, Austria Received July 19, 2005

The study of protein phosphorylation has grown exponentially in recent years, as it became evident that important cellular functions are regulated by phosphorylation and dephosphorylation of proteins on serine, threonine and tyrosine residues. The use of immobilized metal affinity chromatography (IMAC) to enrich phosphopeptides from peptide mixtures has been shown to be useful especially prior to mass spectrometric analysis. For the selective enrichment applying solid-phase extraction (SPE) of phosphorylated peptides, we introduce poly(glycidyl methacrylate/divinylbenzene) (GMD) derivatized with imino-diacetic acid (IDA) and bound FeIII as a material. GMD is rapidly synthesized and the resulting free epoxy groups enable an easy access to further derivatization with, e.g., IDA. Electron microscopy showed that the synthesized GMD-IDA-FeIII for SPE has irregular agglomerates of spherical particles. Inductively coupled plasma (ICP) analysis resulted in a metal capacity of FeIII being 25.4 µmol/mL. To enable on-line preconcentration and desalting in one single step, GMD-IDA-FeIII and Silica C18 were united in one cartridge. Methyl esterification (ME) of free carboxyl groups was carried out to prevent binding of nonphosphorylated peptides to the IMAC function. The recovery for a standard phosphopeptide using this SPE method was determined to be 92%. The suitability of the established system for the selective enrichment and analysis of model proteins phosphorylated at different amino acid residues was evaluated stepwise. After successful enrichment of β-casein deriving phosphopeptides, the established system was extented to the analysis of in vitro phosphorylated proteins, e.g. deriving from glutathione-S-transferase tagged extracellular signal regulated kinase 2 (GST-ERK2). Keywords: glycidyl methacrylate/divinylbenze • immobilized metal affinity chromatography • phosphopeptides • solid-phase extraction • mass spectrometry • GST-ERK2

1. Introduction The reversible phosphorylation of proteins regulates many aspects of cell life. Protein phosphorylation reactions have been clearly established as major regulators of metabolic and signal transduction pathways.1 It is thought that approximately 30% of the proteins encoded by the human genome can be phosphorylated, and abnormal phosphorylation is now recognized as a cause or consequence of many human diseases.2 A number of naturally occurring toxins and tumor promoters exert their effects by targeting particular protein kinases and phosphatases.2 The traditional approach for detecting phosphorylation sites involves radiolabeling of proteins with radioactive 32P marked * To whom correspondence should be addressed. Tel.: ++43/(0) 512/ 507-5195. Fax: ++43/(0) 512/507-2965. E-mail: [email protected]. † Leopold-Franzens University. ‡ Innsbruck Medical University. § Permanent address: Analytical Chemistry Laboratory, Chemistry Department, Faculty of Mathematics and Natural Sciences, Gadjah Mada University, Yogyakarta, Indonesia. E-mail: [email protected].

2312

Journal of Proteome Research 2005, 4, 2312-2319

Published on Web 10/21/2005

phosphate followed by separation using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (PAGE) or chromatographic techniques.3 Although this protocol was widely used because of its high sensitivity based on radioactivity, some obvious disadvantages appeared. Besides being a potential radioactive hazard, the procedure was labor-intensive, prone to sample losses, and often failed to recognize the exact site of phosphorylation.4 Nowadays, several mass spectrometric (MS) techniques, capable of providing sequence information, offer advantages over radiolabeling of proteins. Two complementary ionization techniques, matrix-assisted laser desorption ionization (MALDI), and electrospray ionization (ESI), in combination with a variety of mass analyzers, have been used to identify phosphopeptides and to determine phosphorylation sites.5-7 The advantages of MALDI-TOF MS are the very high sensitivity, the high mass range, and easy operation. One disadvantage of MALDI is the fact that the analysis of compounds with a molecular weight below 600 Da is difficult due to the presence of intense matrix signals. Electrospray ionization (ESI) offers 10.1021/pr050224m CCC: $30.25

 2005 American Chemical Society

research articles

Poly(Glycidyl Methacrylate/Divinylbenzene)-IDA-FeIII

the possibility for on-line coupling with liquid chromatography (LC) or capillary electrophoresis (CE). LC- or CE-ESI-MS offers higher selectivity due to the additional prior separation step. However, ESI has the disadvantage of multiple charging that may complicate the interpretation of data, and the presence of salt, buffers and other additives can reduce the sensitivity significantly.8,9 The mass spectrometric analysis of phosphopeptides, however, is hampered by the electronegativity of the phosphoryl group, resulting in low ionization efficiency, which makes the analysis of multiphosphorylated peptides especially challenging. The main key to succes in phosphoproteome analysis is the selective enrichment of phosphopeptides in the presence of a large number of nonphosphorylated peptides. Nonphosphorylated peptides often reduce or even suppress the ionization of phosphopeptides in the mass spectrometer.7 Immobilized metal ion affinity chromatography (IMAC) has been used for a number of years for the selective enrichment of phosphopeptides from proteolytic digest mixtures containing both phosphorylated and nonphosphorylated components.7,10-22 In this technique, metal ions are bound to a chelating support, FeIII- or GaIII- ions are commonly used for phosphopeptides. Phosphopeptides are selectively bound because of the affinity of the metal ions for the phosphate moiety. The phosphopeptides can be released using high pH or phosphate buffer, the latter usually requiring a further desalting step before LCESI-MS/MS or MALDI-TOF/TOF analysis. Nowadays, a variety of supports are used for IMAC. Agarose supports that are hydrophilic, mechanically and chemically stable, and relatively inert, were the first matrixes used in IMAC.23-28 The common trade names for agarose are Sepharose (Pharmacia) and Bio-Gel A (Bio-Rad).24 Other types of solid supports for IMAC are porous and nonporous silica15,18,29,30 and synthetic polymer-based particles such as cellulose,22,31-33 poly(styrene/divinylbenzene) (PS/DVB),15,34,35 and poly(hydroxy/ methacrylate).36 The three most widely used chelating groups in IMAC are iminodiacetic acid (IDA), tris(carboxymethyl)ethylenediamine (TED), and nitrilotriacetic acid (NTA). IDA is the classical ligand with most IMAC approaches being developed.37 SPE has become one of the most common samples clean up and concentration techniques used in bioanalysis.38 For analytical purposes, SPE is usually performed using a small column or cartridge containing the appropriate packing. Also membranes loaded with appropriate resins, SPME (solid-phase microextraction) and coated capillaries, e.g., for gas chromatographic analysis, have been used for SPE.39 SPE has been used in proteomics for enrichment (preconcentration) and sample clean up (desalting).40-44 In this work, a new polymeric material, GMD-IDA-FeIII, was synthesized and evaluated for the SPE of phosphopeptides. The properties and advantages of this material are discussed in detail by its application upon model proteins as well as for real in vitro phosphorylated protein studies.

2. Materials Acetonitrile (ACN, analytical reagent grade, >99.9%), divinylbenzene (DVB, analytical reagent grade, 80%), methanol (analytical reagent grade, >99.9%), sodium hydroxide (analytical reagent grade, >99%), sodium chloride (analytical reagent grade, g99.5%), sodium carbonate anhydrous (analytical reagent grade, g99.7%), iron(III) chloride hexahydrate (analytical reagent grade, 99.0%), sodium-dihydrogenphosphate, and di-

sodium-hydrogenphosphate were obtained from Merck (Darmstadt, Germany). R,R′-Azobisisobutyronitrile (AIBN, purum, g98%), tris(hydoroxymethyl)aminomethane hydrochloride buffer substance (pH 8.9), ethylenediaminetetraacetic acid (EDTA) disodium salt (g99.0%) and imidazole (99% titration) were obtained from Fluka (Buchs, Switzerland). Myoglobin horse heart (>90%), β-casein bovine (>90%), iminodiacetic acid (IDA, 98%), o-phoshoserine, 2-[N-morpholino]ethanesulfonic acid (MES, 99%), isopropyl-β-thiogalactopyranoside, trizma base (99.9% titration), glutathione (90%), dithiothreitol (98%), and HEPES (99.5% titration) were purchased from Sigma (St. Louis, MO), trifluoroacetic acid (TFA, 98%) from Riedel-de Hae¨n (Hannover, Germany), glycidyl methacrylate (GMA, g97%) and acetyl chloride (98%) were obtained from Aldrich (Milwaukee, WI), sequencing grade-modified trypsin and 50 mM acetic acid buffer from Promega (Madison, WI), phosphopeptides TSTEPQpTQPGENL (m/z ) 1543.5, >97%), pTEEIE (m/z ) 803.7, > 94%), and SFDVPPIDASSPFpSQK (m/z ) 1802, >97%) from Bachem GmbH (Weil am Rhein, Germany), and 2,5dihidroxybenzoic acid (DHB) from Bruker Daltonic GmbH (Bremen, Germany). Ni2+-NTA agarose was obtained from Qiagen (Valencia, CA), GST-sepharose from Amersham Biosciences (Freiburg, Germany). ERK1/2 specific and phosphoERK1/2 specific antibodies were purchased from Cell Signaling (Beverly, MA), His6-specific antibody from BD Biosciences Clontech, (San Jose, CA). Water purified by a Nanopure-Unit (Millipore, Barnstead, MA) was used. GMA and DVB were distilled before use to remove the inhibitor. For preparation of all aqueous solutions, high purity water (Epure, Barnstead Co., Newton, MA) was used. Nitrogen (>99.996%) was from Messer Austria GmbH (Gumpoldskirchen, Austria).

3. Methods Polymerization of Glycidyl Methacrylate/Divinylbenzene (GMD). A 2.5-mL portion of DVB, 120 mL of ACN and 62.5 mg AIBN were added to a 500 mL four-neck round-bottom flask equipped with a mechanical stirrer, condenser, nitrogen inlet and thermometer. The mixture was first purged with nitrogen for 10 min under stirring. With continuous stirring and nitrogen purge the mixture in the flask was heated in an oil bath at 60 °C for 4 h. Then, 5 mL GMA and 100 mg AIBN were added and the mixture was heated at 70 °C for 16 h with continuous stirring under nitrogen. The mixture was cooled to room temperature, filtered and washed thoroughly in a sintered-glass filter with 60 mL of each ACN and methanol. The beads obtained were dried under vacuum at room temperature for 4 h.45 Preparation and Characterization of GMD-Iminodiacetic Acid-IronIII (GMD-IDA-FeIII). GMD was suspended with vigorous stirring in an IDA solution, which was prepared by dissolving 10 g IDA in 2 M Na2CO3 solution (pH 10). Solid NaCl with a final concentration of 2-3% (w/v) and 20 mL methanol were added to get a homogeneous suspension. The reaction mixture was agitated for 4 h at 75 °C. The final reaction product was filtered under vacuum and washed with deionized water until the filtrate turned to neutral. For loading with iron, GMDIDA was mixed with 120 mL of 0.1 M FeCl3 by stirring for 2 h. Then, GMD-IDA-FeIII was vacuum filtered and washed with deionized water until neutral, dried under vacuum at room temperature for 4 h. Metal capacity of FeIII in polymer was measured by inductively coupled plasma optical emission spectroscopy (ICP-OES). The recovery of phosphopeptides in Journal of Proteome Research • Vol. 4, No. 6, 2005 2313

research articles GMD-IDA-FeIII was measured using a phosphorylated model protein digest (β-casein digest). A 100-µL portion of β-casein digest (60 µg/mL) was loaded onto 100 mg of GMD-IDA-FeIII and washed with 1.5 mL of 2-N-morpholinoethanesulfonic acid (MES) (50 mM, pH 6). The phosphopeptides were eluted with 50 mM phosphoserine, whereas 1.5 mL was aliquoted in 10 vials of 150 µL each. For the determination of the recovery, in the first step 5 different concentrations in a range from 0.65 to 65 µg/mL were injected onto the µ-LC-ESI-MS system. The peak area of the monophosphopeptide with the theoretical mass of 2062 Da was calculated. In the second step the same standards were analyzed after SPE using the synthesized GMDIDA-FeIII. Finally, recovery was calculated by comparison of the peak areas obtained from the two independent experiments. Solid-Phase Extraction (SPE). Tryptic Digest of β-Casein and Myoglobin. A 20-µg portion of sequencing grade modified trypsin was dissolved in 200 µL of 50 mM acetic acid buffer and incubated at 37 °C for 30 min. Trypsin solution was added to 1 mg β-casein in 800 µL water, incubated at 37 °C for 16 h. The digest was stopped by adding 25 µL of 2.5% (v/v) TFA and put in the freezer at -20 °C. The same procedure was performed for myoglobin, except it was dissolved in 50 mM tris HCl and denatured at 95 °C for 5 min. Then the mixtures were incubated at 37 °C for 20 h and the digest was stopped by adding 50 µL of 5% (v/v) TFA. Sample Preparation for GST-pERK2. Expression and purification of recombinant proteins as well as in vitro phosphorylation of GST-ERK2 was performed according to ref 22. SPE of Phosphopeptides Originating from β-Casein Digest. A 100-mg portion of GMD-IDA-FeIII was filled into an ISOLUTE filtration cartridge (Grenzach-Wyhlen, Germany), which was sealed at the end with a 20 µm frit (Inula, Vienna, Austria). The cartridge was equilibrated twice with 1 mL of MES (50 mM, pH 6) and loaded with 100 µL of a tryptic digest of β-casein. The catridge was washed 3 times with 1 mL MES (50 mM, pH 6). Phosphopeptides were eluted with 0.5 mL of 50 mM EDTA. SPE of Synthetic Phosphopeptides from a Spiked Myoglobin Digest. Myoglobin digest was spiked with two phosphopeptides TSTEPQpTQPGENL m/z ) 1543.5 (P1) and pTEEIE m/z ) 803.7 (P2) (lowercase-p precedes a phosphorylated residue throughout this work). A 100-µL aliquot of the standard mixture each of 20 pmol/µL was lyophilized and redissolved in 100 µL of 2M methanolic HCl. Esterification was allowed to proceed for 2 h at room temperature.17 The solvent was removed by lyophilization, and the resulting sample was redissolved in 100 µL of 0.1% (v/v) TFA. The SPE cartridge was first packed with 100 mg Si-C18 to enable inline desalting and followed by 100 mg GMD-IDA-FeIII. After loading 100 µL of the sample, nonspecific bound peptides were washed with 100 mM NaCl in 1% (v/v) acetic acid solution. Finally, phosphopeptides were eluted with 50 mM Na2HPO4 (pH 9) from the GMD-IDAFeIII to the Si-C18 material. The phosphopeptides were desalted with 0.5 mL 0.1% (v/v) TFA solution and eluted from the Si-C18 with 0.5 mL 0.1% (v/v) TFA in 50% ACN. SPE from the Mixture of GST-ERK2, GST-pERK2, and Two Synthetic Phosphopeptides. The same procedure as described above was applied for the mixture prepared from 50 µL each of 10 pmol/µL digested GST-ERK2, 5 pmol/µL digested GSTpERK2, 5 pmol/µL phosphopeptide TSTEPQpTQPGENL (m/z ) 1543.5), and 5 pmol/µL phosphopeptide SFDVPPIDASSPFpSQK (m/z ) 1802.0). Mass Spectometry. Samples were analyzed either by MALDITOF/TOF (Ultraflex, Bruker, Bremen, Germany) or micro-(µ)2314

Journal of Proteome Research • Vol. 4, No. 6, 2005

Aprilita et al.

Figure 1. (a) Scanning electron micrograph (SEM) of the GMD polymer, (b) Scheme of synthesis of GMD-IDA-Fe3+.

flow reversed-phase (RP) high-performance liquid chromatography (HPLC) coupled to ESI-MS (Finnigan LCQ ion trap, San Jose, CA). MALDI-TOF/TOF. Sample preparation for MALDI analyses was carried out using dihydroxybenzoic acid (DHB) as a matrix on a steel target. DHB was prepared as a saturated aqueous solution in HPLC grade water. This matrix was mixed in equal volumes with the sample solution. The mixture (1 µL) was pipetted on the target and dried at room temperature. In the case of using 600 µm AnchorChip target (Bruker, Bremen, Germany), 1 µL of the analyte was applied to the anchor, dried and then 0.5 µL DHB (5 mg/mL in H2O) was added. Mass spectra were recorded in positive ion mode. Exclusively positive charged compounds were analyzed and approximately 300 single-shot spectra were accumulated for improved signal-tonoise ratio. Spectra were calibrated externally using Calibration standard I (Bruker). The Flex Analysis version 2.0 and MS Biotools Version 2.2 (Bruker, Bremen, Germany) software packages provided by the manufacturer were used for data processing. µ-LC-ESI-MS of the Phosphorylated/Nonphosphorylated Peptide Mixture after SPE. µ-LC-ESI-MS was performed on a quadrupole ion trap mass spectrometer (Finnigan MAT LCQ, San Jose, CA) equipped with a nano electrosopray ionization source. The monolithic capillary column was connected online to the spray capillary (fused silica, 90 µm o.d., 20 µm i.d.,

research articles

Poly(Glycidyl Methacrylate/Divinylbenzene)-IDA-FeIII

Table 1. Peptides Found in Tryptic Digest of β-Casein mass observed (Da)

sequence range

peptide sequence

742.54 748.46 780.51 830.44 1013.52 1591.93 2061.82 2185.36 2186.11 2432.51 2909.57 3122.33

203-209 108-113 170-176 177-183 106-113 170-183 33-48 184-202 184-202 30-48 184-209 1-25

GPFPIIV EMPFPK VLPVPQK AVPYPQR HKEMPFPK VLPVPQK AVPYPQ FQpSEEQQQTEDELQDK DMPIQAFLLYQEPVLGPVR DMPIQAFLLYQEPVLGPVR IEKFQpSEEQQQTEDELQDKa DMPIQAFLLYQEPVLGPVRGPFPIIV RELEELNVPGEIVEpSLpSpSpSEESITR

a

Figure 2. µ-LC-ESI-MS of elution fraction no.3 from β-casein digest with phosphoserine (inset: SIM-MS of the doubly charged m/z 1032). Chromatographic conditions: Column PS/DVB monolith (60 mm × 200 µm i.d.); mobile phase, A: 0.05% TFA, B: 0.05% TFA in ACN; gradient, 2-70% B in 15 min, 70-100% B in 5 min, 100% B in 1 min, 2% B in 9 min; flow rate, 2.0 µL/min after split, scan, 500-2000 amu in 1.5 s, sample volume, 0.5 µL.

Figure 3. MALDI-TOF mass spectra of β-casein digest. Asterisks indicate not identified peaks. (a) 1 µg/µL β-casein digest before SPE (b) phosphopeptides after SPE originating from 0.2 µg/µL β-casein digest.

Polymicro Technologies, Phoenix, AZ) by means of a microtight union (Upchurch Scientific; Oak Harbor, WA). Nitrogen was used as sheath gas for pneumatically assisted ESI. The mass spectra were recorded on a personal computer equipped with the LCQ Xcalibur software version 1.2 (Finnigan).

Same sequenz as m/z)2061.82 but IEK at the beginning

Figure 4. MALDI-MS/MS spectrum of m/z 2061.9 after SPE using GMD-IDA-FeIII.

The chromatographic separation system consisted of a lowpressure gradient pump (Rheos 2000, Flux Instruments, Basel, Switzerland), an online degasser (Knauer, Berlin, Germany), a microinjector (model 7520, Rheodyne, CA) with a 0.5 µL rotor and a poly-(styrene/divinylbenzene) (PS/DVB) monolith (60 mm × 0.2 mm i.d.) as stationary phase. The mobile phase consisted of 0.05% TFA in water (solvent A) and 0.05% TFA in 70% ACN (solvent B) and was pumped at a constant flow rate of 200 µL/min (actual flow rate 2.0 µL/min after split). Zero Journal of Proteome Research • Vol. 4, No. 6, 2005 2315

research articles

Aprilita et al.

Figure 5. µ-LC-ESI-MS of a myoglobin digest spiked with two phosphopeptides after methyl esterification (1585.5 and 860.1 Da), (a) before SPE and (b) after SPE. Chromatographic conditions: Column PS/DVB monolith (60 mm × 200 µm I. D.); mobile phase, A: 0.05%TFA in water, B: 0.05% TFA in 70% ACN; linear gradient, 0-100% B in 20 min; flow rate, 2.0 µL/min; 60 °C, injection volume 0.5 µL, scan 500-2000 amu in 1.5 s. (c) Corresponding mass spectrum of phosphopeptide P1 and (d) P2 isolated from a myoglobin digest.

time conditions were 100% solvent A, changing from 0 to 20 min to 0% solvent A were carried out at 60 °C.

4. Results and Discussion 4.1. Characterization of GMD-IDA-FeIII Irregular Particles for SPE. The morphology of the synthesized GMD was optically characterized by scanning electron microscopy (SEM). It is shown in Figure 1a that the polymer has irregular agglomerates (20 µm) of spherical particles with a size of approximately 5 µm. Iminodiacetic acid (IDA) was reacted with obtained epoxy groups of GMD under basic conditions. The formed GMD-IDA was washed and then allowed to react with FeCl3 (Figure 1b). Metal capacity of FeIII in GMD-IDA-FeIII was 25.4 µmol/mL as measured by ICP-OES. To test the reproducibility of the established synthetic method, the iron content of 10 independent synthesized GMD-IDA-FeIII charges was determined via ICP-OES and resulted in a RSD of (4.2%. Carrying out mercury intrusion porosimetry (MIP) confirmed that the internal pore size distribution is bimodal with maxima at 300 and 800 Å. The specific surface area was found to be 1.82 m2/g by the Brunnauer Emmet Teller (BET) plot. First experiments with GMD-IDA-FeIII were performed on the basis of a digest of bovine β-casein, a protein with phosphorylation sites on serine residues, by solid-phase extracting the target peptides. The recovery of β-casein using the established system was determined using µ-LC-ESI-MS analysis as described in the Experimental Section. Figure 2 shows the chromatogram with the identified signal at m/z 1032 (M + 2H)2+ used for calculation with the deconvoluted mass of 2062 Da without interfering peaks after SPE with GMD-IDA-FeIII. The recovery was 92% after fractionation. Similar results were obtained earlier using micropellicular silica-IDA-FeIII with a recovery of approximately 94%.18,22 2316

Journal of Proteome Research • Vol. 4, No. 6, 2005

Figure 6. Western blot analysis with antibodies specific for Erk1/ 2, phospho-Erk1/2 and His6 of equal amounts of the phosphorylation mix (+) and of a negative control sample without His6Mek1S/D (-).

4.2. SPE of Phosphopeptides Originating from β-Casein Digest. After successful characterization the specificity of the GMD-IDA-FeIII IMAC material was investigated using bovine β-casein as a model compound due to its commercially availability and its well-characterized five phosphorylation sites at serine residues.14 A MALDI mass spectrum of the β-casein digest before SPE is presented in Figure 3a. Labeled peaks correspond to the matched peptide to β-casein bovine using the MASCOT search engine and the score was significant. It is shown in Table 1, that the found β-casein digest has 5 phosphorylation sites at serine residues and the rest are unphosphorylated peptides. The cartridge was washed with 50 mM MES solution (pH 6) to remove unbound peptides. The phosphopeptides were eluted from the cartridge with 500 µL 50 mM EDTA and analyzed by MALDI-TOF/TOF MS. Three typical phosphopep-

Poly(Glycidyl Methacrylate/Divinylbenzene)-IDA-FeIII

research articles

Figure 7. MALDI-TOF mass spectra of (a) GST-ERK2, (b) GST-pERK2, (c) mixture of GST-ERK2 (10 pmol/µL), GST-pERK2 (5 pmol/µL) and 2 synthetic phosphopeptides22 after methyl esterification, and (d) phosphopeptides from the mixture after SPE.

tide peaks from β-casein digest, m/z 2061.9, 2432.1, and 3122.4, were observed in the MALDI-TOF spectrum depicted in Figure 3b. Thereby, the difference in the signal-to-noise ratio between Figure 3a,b can be explained by the use of 1 µg/µL in Figure 3a and 0.2 µg/µL in Figure 3b. Additionally, in Figure 3a 300 shots, in Figure 3b 75 shots were selected to record the mass spectrum. The m/z 2061.9 from the MALDI-MS spectrum of the elution from the GMD-IDA-FeIII cartridge was chosen for MALDI-TOF/TOF MS. Figure 4 shows the MALDI-TOF/TOF

spectra for the parent ion of m/z 2061.9. The MALDI9MS/MS spectrum of m/z 2061.9 corresponds to MS/MS fragmentation of FQpSEEQQQTEDELQDK. In this spectrum, the presence of abundant [MH - H3PO4]+ or [MH - 98]+ ions and a weaker [MH - HPO3]+ or [MH - 80]+ ion is found. These fragment ions indicate that the peptide is most likely phosphorylated on serine or threonine.46 From the sequence information in Table 1, it can be concluded that m/z 2061.9 has a phosphorylation site on serine. Journal of Proteome Research • Vol. 4, No. 6, 2005 2317

research articles 4.3. SPE of Synthetic Phosphopeptides from a Spiked Myoglobin Digest. For the analysis of a more complex sample, myoglobin was digested with trypsin and then mixed with two phosphopeptides TSTEPQpTQPGENL m/z ) 1543.5 (P1) and pTEEIE m/z ) 803.7 (P2). The resulting mixture contained tryptic peptides and phosphopeptides, each at the 20 pmol/ µL level. To decrease nonspecific binding to the IMAC cartridge, all peptides in the mixture were converted to the corresponding peptide methyl esters.17,47 Aspartic acid, glutamic acid and the C-terminus of each peptide have a static modification of +14 Da (methyl groups) after methyl esterification, the two phosphopeptides showing m/z for P1 ) 1585.5 Da and P2 ) 860.1 Da. For online desalting SPE was done in a “two-dimensional” cartridge containing Si-C18 and GMD-IDA-FeIII as an IMAC phase. The samples before and after SPE were analyzed by µ-LC coupled to MS via an ESI interface. Figure 5 shows the chromatogram of a myoglobin digest spiked with two phosphospoptides before SPE (a) and after SPE (b). As shown in Figure 5, phosphorylated target-peptides could be enriched by the system and the number of nonphosphorylated peptides could be reduced dramatically. Corresponding mass spectra used for identification of the two phosphopeptides P1 and P2 are depicted in Figure 5c and d. 4.4. SPE from the Mixture of GST-ERK2, GST-pERK2, and Synthetic Phosphopeptides. For the analysis of cell lysate deriving samples, the system of GST-ERK2 was investigated. Mitogen activated protein kinase/ERK kinase 1 (MEK1) activates ERK2 by phosphorylation on two amino acid residues, threonine 183 and tyrosine 185. The in vitro phosphorylation of GST-ERK2 was confirmed by Western-blot analysis (Figure 6). The MALDI-TOF mass spectrum of 10 pmol/µL nonphosphorylated GST-ERK2 tryptic digest is shown in Figure 7a and the mass spectrum of 5 pmol/µL phosphorylated GST-pERK2 tryptic digest is shown in Figure 7b. One difference between these two spectra is the mass of 2224 Da only present in Figure 7b. Using Biotools software for data processing, the peptide with the mass of 2224 Da (171-189) was identified as VADPDHDHTGFLTEpYVATR, containing a phosphorylation on tyrosine. The two activating phosphorylation sites, a tyrosine and a threonine (Tyr-185 and Thr-183 of ERK2), are 1 residue apart on the MAP kinases. In vivo and in vitro, phosphorylation of tyrosine precedes phosphorylation of threonine, although phosphorylation of either residue can occur in the absence of the other.48 Figure 7c shows the MALDI-Tof mass spectra without prior enrichment using SPE for a mixture consisting of GST-ERK2, GST-pERK2, and 2 synthetic phosphopeptides after methyl esterification. Synthetic phosphopeptides were added to the mixture as an internal control for the SPE procedure. For identification, in Table 2 the masses of the most interesting fragments before and the most abundant masses after methyl esterification of GST and ERK2 are listed. After SPE applying GMD-IDA-FeIII most of the nonphosphorylated peptides from GST-ERK2 did not appear anymore in the effluent (Figure 7d). Comparison of the mass spectra in Figure 7a-c with the mass spectrum depicted in Figure 7d confirmed the selective enrichment (92%) of phosphorylated peptides. This means that the signal of the phosphopeptides which were suppressed in MALDI-TOF analysis by their nonphosphorylated counterpart without prior enrichment of the phosphorylated peptides became detectable after SPE. The two synthetic phosphopeptides (P1: TSTEPQpTQPGENL, 1585.5 Da 2318

Journal of Proteome Research • Vol. 4, No. 6, 2005

Aprilita et al. Table 2. Comparison of Peptides before and after Methyl Esterification (ME)

no.

sequence

source

m/z that found before ME

1 2 3 4 5 6 7 8

DEGDKWR LTQSMAIIR RIEAIPQIDK AEISMLEGAVLDIR GQVFDVGPR NYLLSLPHK HENIIGINDIIR VADPDHDHTGFLTEpYVATR

GST GST GST GST ERK2 ERK2 ERK2 ERK2

905.4 1032.6 1182.7 1516.9 974.5 1084.7 1406.8 2224.0

m/z that found after ME

961.4 1046.6 1224.6 1572.9 1002.3 1098.4 1448.9 2294.2

and P2: SFDVPPIDASSPFpSQK, 1844.1 Da) and the phosphopeptide P3: VADPDHDHTGFLTEpYVATR (2294.4 Da) from GST-pERK2 could be identified. This shows that all three types of phosphorylation sites, i.e., threonine, serine, and tyrosine are selectively extracted with GMD-IDA-FeIII SPE particles.

5. Concluding Remarks GMD-IDA-FeIII was introduced as a new material for the SPE of phosphopeptides. All three types of phosphorylation peptides i.e., threonine, serine and tyrosine were independently and selectively extracted. GMD-IDA-FeIII revealed high recovery rates (approximately 92%) and high selectivity for the SPE of phosphopeptides from various kind of samples, e.g., β-casein digest, synthetic phosphopeptides from a spiked myoglobin digest and tryptic digest of in vitro phosphorylated protein, GST-ERK2 down to the low picomol level. The choice of washing and eluting detergent did not influence recovery significantly. Using a combination of Si-C18 and GMD-IDAFe3+ in one single cartridge it was also possible to desalt and preconcentrate the sample for MS analysis without disturbing the binding of the phosphopeptides to the IMAC material. GMD-IDA-FeIII is cheap and can be produced very easily. In the future, it can be used for detailed investigations in the field of phosphoproteomics and in high throughput robotic systems for routine analysis. Therefore, the GMD-IDA-FeIII is a competitive alternative to commercially available products that have been used in other IMAC strategies. In a following work, the applicability of this GMD material as a monolithic stationary phase for µ-LC will be demonstrated. Abbreviations. AIBN, R,R′-azobisisobutyronitrile; BET, Brunnauer Emmet Teller; DHB, 2, 5-dihidroxybenzoic acid; DVB, divinylbenzene; GMA, glycidyl methacrylate; GMD-IDA-FeIII, glycidyl methacrylate/divinylbenze-Iminodiacetic acid-iron(III); GST-ERK2, glutathione-S-transferase tagged extracellular signal regulated kinase 2; IDA, iminodiacetic acid; IMAC, immobilized metal affinity chromatography; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight; MIP, mercury intrusion porosimetry; NTA, nitrilotriacetic acid; SPE, solid-phase extraction; SPME, solid-phase microextraction; TED, tris(carboxymethyl)ethylenediamine; TFA, trifluoroacetic acid.

Acknowledgment. Work was supported by the Austrian Genome Program (GEN-AU), Vienna, Austria and by the Special research Program Cell Proliferation and Cell Death in Tumors (SFB021, Austrian Science Fund). Technology Grant Southeast Asia scholarship from The Austrian Federal Ministry for Education, Science and Culture for N.H.A. is also gratefully acknowledged.

research articles

Poly(Glycidyl Methacrylate/Divinylbenzene)-IDA-FeIII

References (1) 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, 225-235. (2) Cohen, P. The origins of protein phosphorylation. Nat. Cell. Biol. 2002, 4, E127-E130. (3) Sulkowski, E. Purification of proteins by IMAC. Trends Biotechnol. 1985, 3, 1-7. (4) Dass, C. Analysis of phosphorylated proteins by mass spectrometry. Appl. Spectrosc. Rev. 2000, 35, 95-128. (5) Neubar, G.; Mann, M. Mapping of Phosphorylation Sites of GelIsolated Proteins by Nanoelectrospray Tandem Mass Spectrometry: Potentials and Limitations. Anal. Chem. 1999, 71, 235-242. (6) Annan, R. S.; Huddleston, M. J.; et al. A Multidimensional Electrospray MS-Based Approach to Phosphopeptide Mapping. Anal. Chem. 2001, 73, 393-404. (7) Raska, C. S.; Parker, C. E.; et al. Direct MALDI-MS/MS of Phosphopeptides Affinity-Bound to Immobilized Metal Ion Affinity Chromatography Beads. Anal. Chem. 2002, 74, 3429-3433. (8) Hop, C. E. C.; Bakhtiar, R. An introduction to electrospray ionization and matrix-assisted laser desorption/ionization mass spectrometry: essential tools in a modern biotechnology environment. Biospectroscopy 1997, 3, 259-280. (9) Carr, S. A.; Hemling, M. E.; et al. Integration of mass spectrometry in analytical biotechnology. Anal. Chem. 1991, 63, 2804-2824. (10) Hart, S. R.; Waterfield, M. D.; et al. Factors governing the solubilization of phosphopeptides retained on ferric NTA IMAC beads and their analysis by MALDI TOFMS. J. Am. Soc. Mass Spectrom. 2002, 13, 1042-1051. (11) Posewitz, M. C.; Tempst, P. Immobilized Gallium(III) Affinity Chromatography of Phosphopeptides. Anal. Chem. 1999, 71, 2883-2892. (12) Li, S.; Dass, C. Iron(III)-Immobilized Metal Ion Affinity Chromatography and Mass Spectrometry for the Purification and Characterization of Synthetic Phosphopeptides. Anal. Biochem. 1999, 270, 9-14. (13) Li, S.; Dass, C. Detection of phosphopeptides in bovine adrenal medulla by using iron (III) immobilized-affinity chromatography and mass spectrometry. EJMS. 1999, 5, 279-284. (14) Zhou, W.; Merrick, B. A.; et al. 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, 273-282. (15) Chaga, G. S. Twenty-five years of immobilized metal ion affinity chromatography: past, present and future. J. Biochem. Biophys. Methods. 2001, 49, 313-334. (16) Stensballe, A.; Andersen, S.; et al. Characterization of phosphoproteins from electrophoretic gels by nanoscale Fe(III) affinity chromatography with off-line mass spectrometry analysis. Proteomics 2001, 1, 207-222. (17) Ficarro, S. B.; McCleland, M. L.; et al. Phosphoproteome analysis by mass spectrometry and its application to Saccharomyces cerevisiae. Nat. Biotechnol. 2002, 20, 301-305. (18) Bonn, G. K.; Kalghatgi, K.; et al. Rapid metal-interaction chromatography of proteins and peptides on micropellicular sorbents. Chromatographia 1990, 30, 484-488. (19) Scanff, P.; Yvon, M.; et al. Immobilized iron affinity chromatographic isolation of phosphopeptides. J. Chromatogr. 1991, 539, 425-432. (20) Thompson, A. J.; Hart, S. R.; et al. Characterization of protein phosphorylation by mass spectrometry using immobilized metal ion affinity chromatography with on-resin β-elimination and Michael addition. Anal. Chem. 2003, 75, 3232-3243. (21) McLachin, D. T.; Chait, B. T. Analysis of phosphorylated proteins and peptides by mass spectrometry. Curr. Opin. Biotechnol. 2001, 5, 591-602. (22) Feuerstein, I.; Morandell, S.; et al. Phosphoproteomic Analysis using Immobilized Metal Ion Affinity Chromatography on the Basis of Cellulose Powder. Proteomics, 2005, 5 (1), 46-54. (23) Porath, J.; Carlsson, J.; et al. Metal chelate affinity chromatography, a new approach to protein fractionation. Nature 1975, 258, 598-599. (24) Muszynska, G.; Andersson, L.; et al. Selective adsorption of phosphoproteins on gel-immobilized ferric chelate. Biochemistry 1986, 25, 6850-6853. (25) Wong, J. W.; Albright, R. L.; et al. Immobilized metal ion affinity chromatography (IMAC)schemistry and bioseparation applications. Sepn. Purif. Methods 1991, 20, 49-106.

(26) Porath, J. Immobilized metal ion affinity chromatography. Prot. Expression. Purif. 1992, 3, 263-281. (27) Muszynska, G.; Dobrowolska, G.; et al. Model studies on iron(III) ion affinity chromatography. II. Interaction of immobilized iron(III) ions with phosphorylated amino acids, peptides and proteins. J. Chromatogr. 1992, 604, 19-28. (28) Zachariou, M.; Traverso, I.; et al. High-performance liquid chromatography of amino acids, peptides and proteins. CXXXI. O-Phosphoserine as a new chelating ligand for use with hard Lewis metal ions in the immobilized-metal affinity chromatography of proteins. J. Chromatogr. 1993, 646, 107-120. (29) El-Rassi, Z.; Horva´th, C. Metal chelate-interaction chromatography of proteins with iminodiacetic acid-bonded stationary phases on silica support. J. Chromatogr. 1986, 359, 241-253. (30) A. Figueroa, C. Corradini, B. Feibush, B. L. Karger, J. Chromatogr. 1986, 371, 335-352. (31) Toomik, R.; Toomik, P. Preparation of ferric adsorbent paper and its interaction with phosphate-containing biomolecules. Prepr. Biochem. 1992, 22, 183-197. (32) Li, Y.; Lingyun, J.; et al. Affinity chromatography of yeast alcohol dehydrogenase using immobilized monochlorotriazine colourless compounds. Sci. China (Series B) 1998, 41, 596-605. (33) Ueda, E. K. M.; Gout, P. W.; et al. Current and prospective applications of metal ion-protein binding. J. Chromatogr. A 2003, 988, 1-23. (34) Shoemaker, M. T.; Haley, B. E. Identification of a guanine binding domain peptide of the GTP binding site of glutamate dehydrogenase: isolation with metal-chelate affinity chromatography. Biochemistry. 1993, 32, 1883-1890. (35) Ficarro, S.; Chertihin, O.; et al. Phosphoproteome Analysis of Capacitated Human Sperm. J. Biol. Chem. 2003, 278, 1157911589. (36) Holmes, L. D.; Schiller, M. R. Immobilized iron(III) metal affinity chromatography for the separation of phosphorylated macromolecules: ligands and applications. J. Liq. Chrom., Relat. Technol. 1997, 20, 123-142. (37) Kastner, M. In Protein Liquid Chromatography; Kastner, M., Ed.; Elsevier: Amsterdam, 2000; pp 301-383. (38) Majors, R. E. New designs and formats in solid-phase extraction sample preparation. LC-GC Eur. 2001, 14, 746-751. (39) Huck, C. W.; Bonn, G. K. Recent developments in polymer-based sorbents for solid-phase extraction. J. Chromatogr. A 2000, 885, 51-72. (40) Bratt, C.; Lindberg, C.; et al. Restricted access chromatographic sample preparation of low mass proteins expressed in human fibroblast cells for proteomics analysis. J. Chromatogr. A 2001, 909, 279-288. (41) Bonneil, E.; Li, J.; et al. Integration of solid-phase extraction membranes for sample multiplexing: application to rapid protein identification from gel-isolated protein extracts. Electrophoresis 2002, 23, 3589-3598. (42) Pang, J. X.;, Ginanni, N.; et al. Biomarker discovery in urine by proteomics. J. Proteome. Res. 2002, 1, 161-169. (43) Janini, G. M.; Conrads, T. P.; et al. Development of a twodimensional protein-peptide separation protocol for comprehensive proteome measurements. J. Chromatogr. B 2003, 787, 43-51. (44) Shen, Y.; Toli, N.; et al. Ultrasensitive Proteomics Using HighEfficiency On-Line Micro-SPE-NanoLC-NanoESI MS and MS/ MS. Anal. Chem. 2004, 76, 144-154. (45) Zhang, S.; Huang, X.; et al. Preparation of monodisperse porous polymethacrylate microspheres and their application in the capillary electrochromatography of macrolide antibiotics. J. Chromatogr. A 2002, 948, 193-201. (46) Annan, R. S.; Carr, S. A. Phosphopeptide Analysis by MatrixAssisted Laser Desorption Time-of-Flight Mass Spectrometry. Anal. Chem. 1996, 68, 3413-3421. (47) Trojer, L.; Stecher, G.; et al. Characterisation and evaluation of metal-loaded iminodiacetic acid-silica of different porosity for the selective enrichment of phosphopeptides. J. Chromatogr. A 2005, 1079, 197-207. (48) Cobb, M. H.; Goldsmith, E. J. How MAP kinases are regulated. J. Biol. Chem 1995, 270, 14843-14846.

PR050224M Journal of Proteome Research • Vol. 4, No. 6, 2005 2319