Phosphopeptide Modification and Enrichment by Oxidation–Reduction

LETTER

Phosphopeptide Modification and Enrichment by Oxidation–Reduction Condensation Mangalika Warthaka, Paulina Karwowska-Desaulniers, and Mary Kay H. Pflum* Department of Chemistry, Wayne State University, 5101 Cass Avenue, Detroit, Michigan 48202

P

hosphorylation is an important post-translational modification that influences protein function and cell signaling (1). Because of its role in regulating protein activity, characterizing and monitoring phosphorylation events is a fruitful area of proteomics research. Although traditional methods for monitoring phosphorylated proteins include in vivo 32P-labeling and gel electrophoresis, mass spectrometric (MS) analysis has emerged as a useful tool for phosphoproteomics applications (2). To increase the probability of detecting phosphopeptides in complex peptide mixtures, MS analysis is typically preceded by a phosphopeptide enrichment step, such as immobilized metal affinity chromatography (IMAC) (2, 3). Unfortunately, the reliance of IMAC methods on noncovalent interactions for phosphopeptide capture results in contamination with unphosphorylated peptides (4). Towards more effective enrichment strategies, phosphopeptides have been attached to a solid phase via covalent bonds. For example, multiple strategies rely on phosphate ␤-elimination followed by thiol conjugate addition for covalent solidphase capture (4–9). After the solid phase is washed, phosphopeptides are recovered by cleavage from the resin. A limitation of the strategy is that ␤-elimination occurs with unphosphorylated peptides, including serine, threonine, and glycosylated amino acids (10, 11), which contaminate the enriched mixture and complicate the MS analysis. In addition, only phosphoserineand phosphothreonine-containing peptides are susceptible to ␤-elimination conditions,

www.acschemicalbiology.org

which excludes analysis of phosphotyrosinecontaining peptides. As an alternative to ␤-elimination strategies, two direct phosphate modification reactions have been previously explored for phosphopeptide capture. In the first approach, a phosphotyrosine-containing peptide was alkylated using an ␣-diazo carbonyl functionalized resin for capture and release (12). In the second approach, phosphopeptides from cell lysates were enriched using carbodiimide condensation (13, 14). Like the ␤-elimination approach, a limitation of the two covalent chemistries is that their compatibility with phosphoserine, phosphothreonine, and phosphotyrosine peptides has not been established. Specifically, the alkylation chemistry has only been validated with phosphotyrosine-containing peptides, while the carbodiimide condensation strategy required initial enrichment of phosphotyrosine-containing peptides using antibodies. Because covalent phosphopeptide enrichment has the potential to significantly aid phosphoproteomics research, new strategies that enrich all phosphopeptides and avoid contamination by unphosphorylated peptides are needed. In search of a new chemical strategy to enhance the chemical selectivity of phosphopeptide capture, we considered the oxidation–reduction condensation, where a phosphine and disulfide act to activate a phosphate for subsequent coupling with an alcohol or amine (Scheme 1). Oxidation– reduction condensation was originally reported for alkylation of phosphates to

A B S T R A C T Many cellular processes are regulated by the reversible phosphorylation of proteins. Despite the importance of monitoring protein phosphorylation, available methods to modify and enrich phosphopeptides from complex mixtures for subsequent mass spectrometric analysis are challenging. Here the oxidation–reduction condensation was shown for the first time to directly modify the phosphate of phosphopeptides and phosphoproteins. By coupling with a solid-phase resin, the oxidation–reduction condensation was validated for capture and recovery of phosphoserine-, phosphothreonine-, and phosphotyrosine-containing peptides from a peptide mixture. In addition, full-length phosphoproteins or phosphopeptides from a protein digestion were captured and recovered using the oxidation–reduction condensation, demonstrating its compatibility with protein mixtures. The strategy modifies all phosphopeptides, maintains high chemical selectivity, requires only two steps, and relies on commercially available reagents, suggesting that the oxidation–reduction condensation has the potential to enhance phosphopeptide enrichment methods and encourage development of efficient biochemical and proteomics tools targeting phosphorylation.

*Corresponding author, pfl[email protected]. Received for review August 16, 2006 and accepted October 22, 2006. Published online December 1, 2006 10.1021/cb6003564 CCC: $33.50 © 2006 by American Chemical Society

VOL.1 NO.11 • ACS CHEMICAL BIOLOGY

697

OH O P OH O AcLRRA

N H

L O

N H

OH O P OH O

HCl, MeOH 0.5 h, RT OH

AcLRRA

O

N H

1

L O

N H

H2N OMe O

O− O P NH O

R

PPh3, PySSPy DIPEA, DMF 5 h, RT

AcLRRA

N H

1b

L O

N H

R OMe O

1c

Scheme 1. The two-step protection/oxidation–reduction condensation of phosphoserine-containing peptide 1 with benzyl amine.

sized to assess chemical selectivity. Because the carboxyl terminus and carboxylic acid side chain of each peptide would react under conditions of oxidation–reduction condensation, the acids were selectively protected in the presence of phosphate esters, as previously reported (Scheme 1) (12, 14). MALDI-TOFMS data of the peptide products after protection were consistent with conversion of the carboxyl termini of peptides 1–3 to methyl esters (Table 1). In addition, aspartic acid-containing peptide 4 displayed a mass consistent with protection of the carboxyl terminus and carboxylic acid side chain. The protected peptides were used in the oxidation– reduction condensation reactions without purification. The condensation of peptides 1–4 was initially explored with benzyl amine (Scheme 1, where R ⫽ Ph) because earlier work demonstrated an 80% conversion of phosphate to phosphoramidates (15). The oxidation– reduction condensation was carried out with the protected phosphoserineTABLE 1. MALDI-TOFMS data of peptides 1–8 after containing carboxylic acid protection and oxidation–reduction peptide 1b condensation with benzylamine (Scheme 1), and MALDI-TOFMS Protected peptides Condensation products Peptides data (Table 1) a b a b Calcd Obsd Calcd Obsd indicated conver1 (AcLRRApSLG) 908.46 908.47 997.53 997.58 sion to a single 2 (AcLRRASLG) 828.50 828.47 917.56 828.47d product consistent 3 (AcLRRACLG) 844.47 844.51 917.56 844.51d with benzylamine 4 (AcARRADLG) 828.46c 828.44 903.56 828.44d condensation 5 (AcLRRApTLG) 922.45 922.55 1011.54 1011.61 (1c). In contrast, 6 (AcLRRATLG) 842.52 842.52 931.58 842.52d unphosphory7 (AcLRRApYLG) 984.49 984.46 1073.54 1073.68 lated peptides 2, 8 (AcARRAYLG) 862.48 862.71 951.57 862.71d 3, and 4 were a unreactive under Calculated values are masses [M ⫹ H]⫹ of protected peptides and expected condensation products. bObserved MALDI-TOFMS data were the condensation collected in the positive ion mode. cCalculated mass of peptide proconditions; only tected at the C-terminus and aspartic acid side chain. dMass of prothe presence of tected peptide, indicating no reaction under the condensation the starting proconditions. tected peptide

create mixed phosphate diesters and phosphoramidates (15). Subsequently, oxidation–reduction condensation of nucleic acids was exploited to create various phosphate-modified DNA and RNA analogues (16–18). In addition, the oxidation– reduction condensation was employed for amino acid coupling in peptide synthesis (19). Despite the reported efficiency for phosphate modification and compatibility with peptides, the oxidation–reduction condensation has not been exploited with phosphopeptides. We initially tested the oxidation–reduction condensation for phosphopeptide modification using synthetic peptides to allow a rigorous assessment of chemical selectivity. Phosphokemptide, (peptide 1, Table 1), a natural enzymatic product of cAMP-dependent protein kinase (20), was generated. In addition, peptides where the phosphoserine of phosphokemptide was replaced by serine, cysteine, or aspartic acid (peptides 2, 3, and 4, Table 1) were synthe-

698

VOL.1 NO.11 • 697–701 • 2006

WARTHAKA ET AL.

was observed (Table 1). The data indicate that oxidation–reduction condensation with benzylamine is dependent on the presence of a phosphate group. To assess the efficiency of the two-step phosphate modification reaction, the product mixture after oxidation–reduction condensation of peptide 1b with benzyl amine was separated using reverse-phase HPLC (Supplementary Figure 1). While starting protected peptide 1b was present in the reaction mixture, the predominant peptide peak corresponded to the condensation product 1c. Quantification of the peptide peaks indicated an average of 89% conversion to condensation product. To test the reactivity of the other phosphorylated peptides under condensation conditions, we synthesized peptides 5, 6, 7, and 8 by replacing the phosphoserine of peptide 1 with phosphothreonine, threonine, phosphotyrosine, and tyrosine, respectively (Table 1). Oxidation–reduction condensation was carried out with the protected peptides, and MALDI-TOFMS data of the reaction products indicated that only the phosphopeptides were modified under the reaction conditions; unphosphorylated peptides were unreactive. HPLC analysis indicated an average of 90% conversion of the phosphothreonine peptide 5 to condensation product (Supplementary Figure 2), demonstrating that the condensation reaction equally modifies phosphoserine- and phosphothreonine-containing peptides. In the case of phosphotyrosine peptide 7, HPLC analysis revealed an average of 58% conversion to condensation product (Supplementary Figure 3). In total, the data indicate that the oxidation–reduction condensation reaction modifies all phosphatecontaining peptides, with phosphoserine and phosphothreonine modified to a greater extent. In addition, phosphopeptides but not unphosphorylated peptides are susceptible to the reaction conditions. www.acschemicalbiology.org

LETTER 908.50

984.47

1324.78

1406.87 1380.65

922.46 984.46

842.93 828.66 800

m/z

1600

800

The next step toward development of a selective phosphopeptide enrichment strategy was testing the oxidation–reduction condensation for solid-phase phosphopeptide capture. A mixture of synthetic peptides was employed initially to assess the compatibility and selectivity of the condensation in solid-phase capture and release. Phosphopeptides 1, 5, and 7 were subjected to the two-step reaction conditions in the presence of glycine-conjugated Wang resin (Scheme 1, where R ⫽ glycine-Wang resin) to test chemical compatibility. To simultaneously assess chemical selectivity, six competing unphosphorylated peptides were also included in the peptide mixture: peptides 2, 6, and 8, angiotensin, glucagon, and an Abl substrate peptide (Table 1 and Figure 1). These unphosphorylated peptides were selected because they contain a random pool of naturally occurring amino acids. After solid-phase capture and resin washing, the bound peptides were released under acidic conditions. Whereas all nine peptides were observed as a mixture in solution before solid-phase capture (Figure 1, panel a), only the three phosphopeptides were recovered after release from the resin (Figure 1, panel b). Significantly, no trace of unphosphorylated peptide was detected after recovery. The data indicate that the two-step protection/condensation reaction strategy successfully captures all three phosphate-containing peptides in the presence of possible contaminating unphosphorylated peptides. In addition, the fact that the six unphosphorylated pepwww.acschemicalbiology.org

922.48

b 862.80 908.49

a

m/z

1600

Figure 1. Enrichment of phosphopeptides from a peptide mixture. a, b) MALDI-TOFMS analysis of phosphopeptides 1, 5, and 7 (shown in red) captured on a solid phase in the presence of six unphosphorylated peptides: peptides 2, 6, and 8, angiotensin (DRVYIHPFHL, 1324.78 m/z), glucagon (AQDFVQWLMNT, 1380.65 m/z), and Abl substrate peptide (AcEAIYAAPFAKKK, 1406.87 m/z). Shown are reaction mixtures before solid-phase capture (a) and after cleavage from the resin (b).

tides contained common functional groups present in all proteins—carboxylic acids, alcohols, thioethers, amines, amides, guanidines, imidazoles, and indoles—indicates that the reaction strategy is selective for modification and capture of phosphatecontaining peptides. To assess the limits of detection with the solid-phase capture and recovery by oxidation–reduction condensation, decreasing quantities of phosphoserine peptide 1 were subjected to bead binding and cleavage. The phosphopeptide was recovered and identified by MALDI-TOFMS with starting quantities as little as 100 pmol (data not shown). Because detection limits as low as 100 pmol have proven reliable for capture of phosphopeptides in cell lysates (14), the data indicate that the oxidation–reduction condensation is compatible with the concentrations of phosphopeptides in biological samples. With the high selectivity and limits of detection of the oxidation–reduction condensation established using synthetic peptides, we sought to validate the reaction for solid-phase capture of a phosphopeptide from a full-length protein digestion. The ␤-casein protein, which contains a phosphoserine residue, was used previously to validate phosphopeptide enrichment for proteomics applications (13, 14). Similarly, the two-step oxidation–reduction condensation was performed with a trypsin digestion of the ␤-casein protein. Full-length ␤-casein was initially reduced and alkylated to improve solubility in organic solvents

before digestion with trypsin. Prior to enrichment, the digested mixture showed multiple peptide fragments, with only a small peak corresponding to the phosphopeptide (Figure 2, panel a). The peptide fragments were subjected to the two-step capture with the glycine-conjugated Wang resin. Following cleavage from the resin, the only peptide recovered corresponded to the phosphorylated peptide (Figure 2, panel b). The data with ␤-casein indicate that the oxidation– reduction condensation is compatible with peptide mixtures derived from a phosphoprotein for application to proteomics research. Previous phosphopeptide enrichment using the carbodiimide condensation demonstrated 20–35% recovery of phosphopeptides from a ␤-casein digestion (13, 14). To directly compare the carbodiimide and oxidation–reduction condensation chemistries, we used quantitative MS analysis to assess the efficiency of the oxidation–reduction condensation for phosphopeptide capture on a solid phase, as previously reported (14). Quantitative MS indicated that an average of 37% of the phosphopeptide was recovered from the solid phase (Supplementary Figure 4), indicating that the phosphopeptide recovery with the oxidation–reduction condensation is comparable to the previous carbodiimide strategy. Coupled with the proven high selectivity and phosphopeptide compatibility, the experiments with ␤-casein establish that the oxidation–reduction condensation provides a needed alternative to previous chemical approaches. VOL.1 NO.11 • 697–701 • 2006

699

2228.81 m/z

1

2

3

β-Casein

2077.77

2372.82

2160.46

2277.86

1802.42

1540.31

1295.97

1179.81 1000

c

2160.09

b

1412.12

a

d

4

5

6

CREB 2500

1000

m/z

2500

The spectrum of phosphopeptides identified Figure 2. Enrichment of phosphopeptides from a protein digestion or full-length phosphoproteins. a, b) MALDIfrom a cellular mixture is TOFMS of trypsin digested ␤-casein before (a) and after (b) solid-phase phosphopeptide capture and release. The dependent on the type of protected phosphopeptide (FQSEEQQQTEDELQDK, 2160.09 m/z) is minimally deprotected (2077.77 m/z) under the enrichment chemistry cleavage conditions (shown in red). c) Full-length ␤-casein (lane 1) was recovered either untreated (lane 2) or after phosphatase treatment (lane 3). d) Full-length CREB partially purified after over-expression in bacteria (lane 4) was employed (24). For captured and recovered either untreated (lane 5) or after PKA phosphorylation (lane 6). The faster-migrating band example, different protobeneath CREB corresponds to autophosphorylated PKA. cols for IMAC enrichment identified overlapping but nonidentical To further assess the chemical versatility capture (Figure 2, panel d, lane 5). In addiphosphopeptides from mouse synaptic protion, a protein band migrating at the same of the oxidation–reduction condensation, molecular weight as PKA was also observed teins (24). In addition, ␤-elimination and we tested the capture and recovery of fullIMAC methods yielded a nonredundant (Figure 2, panel d, lane 6), indicating the length phosphoproteins. Full-length ␤group of phosphopeptides from mouse capture of autophosphorylated PKA (22, casein (Figure 2, panel c, lane 1) was incubrain (4). These results suggest that multiple 23). The experiments with full-length CREB bated with glycine-conjugated Wang resin under conditions of the oxidation–reduction and PKA validate the compatibility of phos- phosphopeptide enrichment approaches are needed to fully characterize the phosphoprotein enrichment with multiple procondensation. Following cleavage from the teins. The fact that CREB maintains a single phoproteome. The oxidation–reduction solid phase, the phosphoprotein was recondensation described here provides a phosphorylated serine (21) indicates that covered and visualized by silver staining general phosphopeptide enrichment the solid-phase capture method effectively (Figure 2, panel c, lane 2). Importantly, approach with proven compatibility with enriches proteins with only one phosunphosphorylated ␤-casein generated by phorylated amino acid. Importantly, phos- phosphoserine, phosphothreonine, and incubation with calf intestinal phosphatase was not recovered (Figure 2, panel c, lane 3), phorylated CREB and PKA were enriched from phosphotyrosine peptides. In addition, its high chemical selectivity will avoid cona mixture of proteins derived from cell indicating that the capture is phosphatetamination by unphosphorylated peptides, lysates, suggesting that the chemistry is dependent. Because no direct phosphate modification chemistry has been previously appropriate for phosphoproteomics applica- including glycosylated peptides (10, 11), which will simplify MS analysis. Because tions. shown to enrich full-length proteins, the We hypothesized that use of unexplored covalent phosphopeptide enrichment oxidation–reduction condensation provides a general method for solid-phase capture of chemistry for phosphopeptide modification methods have the potential to significantly aid phosphoprotein analysis, new would have the potential to enhance the phosphoproteins in addition to phosphostrategies like the oxidation–reduction compatibility and selectivity of enrichment peptides. strategies. Here we demonstrate for the first condensation will encourage developFinally, to demonstrate general compattime that an oxidation–reduction condensa- ment of efficient biochemical and proibility with multiple full-length proteins, the teomics tools targeting protein phostion reaction selectively converts the phosoxidation–reduction condensation was phorylation. tested with the cyclic adenosine monophos- phate of phosphopeptides and phosphoproteins into phosphoramidates, allowing phate response element binding (CREB) protein, which is phosphorylated by protein solid-phase enrichment. The highlights of METHODS the strategy include high chemical selectivkinase A (PKA) at Ser133 (21). Full-length Phosphopeptide Modification Reactions. Pepity for phosphorylated peptides and proteins; tides 1–8 were synthesized using standard solidCREB was over-expressed in bacteria as a hexa-his tagged fusion protein and partially compatibility with phosphoserine, phospho- phase fmoc chemistry (see Supporting Information). HPLC-purified peptide (0.04 ␮mol) was resusthreonine, or phosphotyrosine; and recovery pended in 0.2 mL of anhydrous MeOH, and acetyl purified using Ni-nitrilotriacetic (NTA) resin chloride (2 N, 0.4 mmol) was added to generate (Figure 2, panel d, lane 4). The CREB protein of full-length phosphoproteins, distinguishanhydrous HCl. After stirring for 30 min and solvent ing it from previous phosphate modification was untreated or phosphorylated with PKA evaporation, the protected peptide precipitate was strategies. In total, the data presented here before capture and recovery via the protecresuspended in 0.3 mL of anhydrous DMF under argon gas, and 2,2=-dithiodipyridine (0.4 ␮mol, demonstrate that unexplored phosphate tion/oxidation–reduction condensation. 10 equiv), N,N⬘-diisopropylethylamine (DIPEA) CREB was enriched from the protein mixture modification chemistry such as the oxida(0.04 ␮mol, 1 eq), triphenylphosphine (0.4 ␮mol, tion–reduction condensation has the poten- 10 equiv), and benzyl amine (0.6 ␮mol, 15 equiv) in a phosphorylation-dependent manner were added with vigorous shaking. After 5 h, the tial to enhance phosphopeptide and phos(Figure 2, panel d, lane 6); unphosphoryreaction mixture was analyzed by MS and/or HPLC (see Supporting Information). phoprotein enrichment strategies. lated CREB was absent after solid-phase 700

VOL.1 NO.11 • 697–701 • 2006

WARTHAKA ET AL.

www.acschemicalbiology.org

LETTER Capture of Phosphopeptides Using GlycinePreloaded Wang Resin. The carboxylic acid groups of the peptide (0.04 ␮mol) or peptide mixture (0.32 ␮mol; peptides 1, 2, 5, 6, 7, and 8, angiotensin, glucagons, and Abl kinase substrate peptide in a 1:1:1:1:1:1:3:3:2 molar ratio) were protected as described, and the peptide precipitate was resuspended in 0.1 mL of dry DMF. Fmoc-deprotected glycine-preloaded Wang resin (20 mg) that was presoaked in dry DMF (0.2 mL) was added under argon. The oxidation–reduction condensation reaction was initiated by addition of the reagents described above, with exception of benzyl amine. After shaking for 14 h, the resin was collected and washed with DMF (10⫻), dichloromethane (4⫻), 50% dichloromethane in ether (3⫻), and ether (3⫻). The washed beads were incubated with a mixture of 95% TFA, 2.5% TIS, and 2.5% water (0.4 mL) for 2 h to cleave the peptide. After solvent evaporation, the remaining peptide precipitate was resuspended in water before analysis by MALDI-TOFMS (see Supporting Information). Trypsin Digestion of ␤-Casein. Bovine ␤-casein from milk (0.03 ␮mol, Sigma) was resuspended in trypsin buffer (50 mM ammonium carbonate, pH 8), reduced with 5 mM DTT at 60 °C for 30 min, and alkylated with 15 mM iodoacetamide at 25 °C for 30 min in the dark. Sequencing-grade trypsin (Promega) was added in 1:100 ratio (trypsin/protein, w/w) at 37 °C overnight. After lyophilization, the peptide fragments were protected in a solution of 100 ␮L of anhydrous MeOH and 500 ␮L acetyl chloride with vigorous shaking for 2 h at 12 °C. Samples were lyophilized, and the oxidation–reduction condensation was initiated by addition of 10 mg of Fmoc-deprotected glycine-preloaded Wang bead resin (presoaked and deprotected as described previously), 2,2=-dithiodipyridine (302 mM), triphenylphosphine (130 mM), and DIPEA (38 mM) in 0.3 mL of dry DMF for 15 h under argon, followed by washing of the beads and cleavage as described. MALDI-TOF analysis is described in the Supporting Information. Capture of Full-Length Proteins. Full-length CREB was overexpressed in BL21 E. coli cells transformed with the T7-7 CREB plasmid, a kind gift of the Montminy lab (25), before purification with Ni-NTA (see Supporting Information). To phosphorylate the CREB-containing mixture, 0.014 ␮mol of total protein was incubated with 1500 units of PKA and 200 ␮M ATP in 1⫻ kinase reaction buffer. To produce unphosphorylated ␤-casein protein, calf intestinal phosphatase (1 ␮L) was incubated with 0.03 ␮mol of protein for 1 h at 30 °C. The CREB mixture (0.014 ␮mol) or ␤-casein protein (0.03 ␮mol) was used directly in solid-phase capture and release as described. After evaporation of the solvent, the proteins were separated using 12% SDS-PAGE before visualization by silver staining. Acknowledgment Funding was provided by Research Corp. (Grant RI1006) and Wayne State University. We thank M. Montminy for the T7-7 CREB plasmid and E. Aubie, A. Bieliauskas, J. Flammer, K. Green, and S. Suwal for helpful comments.

www.acschemicalbiology.org

Supporting Information Available: This material is available free of charge via the Internet.

REFERENCES 1. Manning, G., Whyte, D. B., Martinez, R., Hunter, T., and Sudarsanam, S. (2002) The protein kinase complement of the human genome, Science 298, 1912–1926. 2. Garcia, B. A., Shabanowitz, J., and Hunt, D. F. (2005) Analysis of protein phosphorylation by mass spectrometry, Methods 35, 256–264. 3. Corthals, G. L., Aebersold, R., and Goodlett, D. R. (2005) Identification of phosphorylation sites using microimmobilized metal affinity chromatography, Mass spectrometry: Modified proteins and glycoconjugates, Methods in Enzymology, Vol. 405, pp 66 –81, Academic Press, Boston. 4. Tseng, H., Ovaa, H., Wei, N. J. C., Ploegh, H., and Tsai, L. (2005) Phosphoproteomic analysis with a solidphase capture-release-tag approach, Chem. Biol. 12, 769–777. 5. Oda, Y., Nagasu, T. T., and Chait, B. T. (2001) Enrichment analysis of phosphorylated proteins as a tool for probing the phosphoproteome, Nat. Biotechnol. 19, 379–382. 6. Knight, Z. A., Schilling, B., Row, R. H., Kenski, D. M., Gibson, B. W., and Shokat, K. M. (2003) Phosphospecific proteolysis for mapping sites of protein phosphorylation, Nat. Biotechnol. 21, 1047–1054. 7. Goshe, M. B., Veenstra, T. D., Panisko, E. A., Conrads, T. P., Angell, N. H., and Smith, R. D. (2002) Phosphoprotein isotope-coded affinity tags: Application to the enrichment and identification of lowabundance phosphoproteins, Anal. Chem. 74, 607–616. 8. Adamczyk, M., Gebler, J. C., and Wu, J. (2001) Selective analysis of phosphopeptides within a protein mixture by chemical modification, reversible biotinylation and mass spectrometry, Rapid Commun. Mass. Spectrom. 15, 1481–1488. 9. Rusnak, F., Zhou, J., and Hathaway, G. M. (2002) Identification of phosphorylated and glycosylated sites in peptides by chemically targeted proteolysis, J. Biomol. Tech. 13, 228–237. 10. Li, W., Backlund, P. S., Boykins, R. A., Wang, G., and Chen, H. (2003) Susceptibility of the hydroxyl groups in serine and threonine to ␤-elimination/Michael addition under commonly used moderately hightemperature conditions, Anal. Biochem. 323, 94–102. 11. Rusnak, F., Zhou, J., and Hathaway, G. M. (2004) Reaction of phosphorylated and O-glycosylated peptides by chemically targeted identification at ambient temperature, J. Biomol. Tech. 15, 296–304. 12. Lansdell, T. A., and Tepe, J. J. (2004) Isolation of phosphopeptides using solid phase enrichment, Tetrahedron Lett. 45, 91–93. 13. Zhou, H., Watts, J., and Aebersold, R. (2001) A systematic approach to the analysis of protein phosphorylation, Nat. Biotechnol. 19, 375–378. 14. Tao, W. A., Wollscheid, B., O’Brien, R., Eng, J. K., Li, X.-j, Bodenmiller, B., Watts, J. D., Hood, L., and Aebersold, R. (2005) Quantitative phosphoproteome analysis using a dendrimer conjugation chemistry and tandem mass spectrometry, Nat. Methods 2, 591–598.

15. Mukaiyama, T., and Hashimoto, M. (1971) Phosphorylation of alcohols and phosphates by oxidation–reduction condensation, Bull. Chem. Soc. Jpn. 44, 196–199. 16. Mukaiyama, T., and Hashimoto, M. (1972) Synthesis of oligothymidylates and nucleoside cyclic phosphates by oxidation–reduction condensation, J. Am. Chem. Soc. 94, 8528–8532. 17. Stepinski, J., Waddell, C., Stolarski, R., Darzynkiewicz, E., and Rhoads, R. E. (2001) Synthesis and properties of mRNAs containing the novel “antireverse” cap analogs 7-methyl(3=-O-methyl)GpppG and 7-methyl (3=-deoxy)GpppG, RNA 7, 1486–1495. 18. Prabahar, K. J., and Ferris, J. P. (1997) Adenine derivatives as phosphate-activating groups for the regioselective formation of 3=,5=-linked oligoadenylates on montmorillonite: Possible phosphate-activating groups for the prebiotic synthesis of RNA, J. Am. Chem. Soc. 119, 4330–4337. 19. Mukaiyama, T., Matsueda, R., and Suzuki, M. (1970) Peptide synthesis via the oxidation–reduction condensation by the use of 2,2=-dipyridyldisulfide as an oxidant, Tetrahedron Lett. 11, 1901–1904. 20. Meyer, H. E., Hoffmann-Posorske, E., Korte, H., and Heilmeyer, L. M. (1986) Sequence analysis of phosphoserine-containing peptides: modification for picomolar sensitivity, FEBS Lett. 204, 61–66. 21. Gonzalez, G. A., and Montminy, M. R. (1989) Cyclic AMP stimulates somatostatin gene transcription by phosphorylation of CREB at serine 133, Cell 59, 675–680. 22. Yonemoto, W., Garrod, S. M., Bell, S. M., and Taylor, S. S. (1993) Identification of phosphorylation sites in the recombinant catalytic subunit of cAMPdependent protein kinase, J. Biol. Chem. 268, 18626–18632. 23. Steinberg, R. A., Cauthron, R. D., Symcox, M. M., and Shuntoh, H. (1993) Autoactivation of catalytic (C alpha) subunit of cyclic AMP-dependent protein kinase by phosphorylation of threonine 197, Mol. Cell. Biol. 13, 2332–2341. 24. Collins, M. O., Yu, L., Coba, M. P., Husi, H., Campuzano, I., Blackstock, W. P., Choudhary, J. S., and Grant, S. G. N. (2005) Proteomic analysis of in vivo phosphorylated synaptic proteins, J. Biol. Chem. 280, 5972–5982. 25. Parker, D., Ferreri, K., Nakajima, T., LaMorte, V. J., Evans, R., Koerber, S. C., Hoeger, C., and Montminy, M. R. (1996) Phosphorylation of CREB at Ser-133 induces complex formation with CREB- binding protein via a direct mechanism, Mol. Cell. Biol. 16, 694–703.

VOL.1 NO.11 • 697–701 • 2006

701