Phosphopeptide Extraction after Electrophoresis and Analysis with MALDI-MS.

A. G. Craig and W. H. Fischer

The Clayton Foundation Laboratories for Peptide Biology, The Salk Institute, LaJolla, CA, U.S.A.



Abstract
Phosphorylated kemptide (pkemptide) was recovered from a cellulose plate following electrophoresis and the extract analyzed with matrix assisted laser desorption/ionization mass spectrometry (MALDI-MS). Four different amounts of pkemptide (which included a trace amount of 32P pkemptide) ranging from 1 ng to 20 ng were spotted on a cellulose plate and submitted to electrophoresis under acidic conditions (pH 1.9). The portion of the cellulose corresponding to the radioactive peptide was removed from the plate and the phospho peptide extracted and successfully analyzed with MALDI-MS in all cases.

Introduction
Phosphorylation of proteins is a dominant post-translational event in cellular regulation. Many signal transduction pathways include cascades of phosphorylation events (1, 2) . Phosphorylation is also prominently involved at multiple levels of transcriptional regulation (3) and cell cycle control (4) . Transformation of cells is often accompanied by changes in the phosphorylation status of key proteins. Amino acid residues that serve as phospho-acceptors include tyrosine, serine, threonine and to a lesser extent histidine. The enzymes involved in phosphorylation can be classified into two broad classes: tyrosine kinases and serine/threonine kinases. The identification of the site of modification is a prerequisite for understanding the role of phosphorylation in the modulation of kinase substrate activities. In this study we explore the sensitivity of matrix assisted laser desorption/ionization mass spectrometry (MALDI-MS) for identifying phosphopeptides after transfer from cellulose 2-D electrophoresis plates.

Results
We chose the model peptide, kemptide (LRRASLG) since it is readily phosphorylated at serine by cAMP dependent protein kinase (PK-A). A trace amount of radioactively labeled peptide was mixed with the non-radioactively labeled peptide to identify the region on the cellulose plate where the non radioactive labeled phosphopeptide had migrated.
The positive MALD mass spectrum of a 0.1 ng ("100 fmol) sample of kemptide analyzed directly (i.e. without spotting on a cellulose plate and extraction or electrophoresis) resulted in an intense signal. In contrast, when 0.1 ng of kemptide was spotted onto a cellulose plate and extracted (without electrophoresis) no signal was observed. When taking into account that the aliquot applied to the thin layer of matrix corresponds to approximately 10% of the decant, this result is not so surprising! Intact molecule ions were observed when 1 and 10 ng of kemptide respectively were spotted onto a cellulose plate and recovered using the procedure described. In these mass spectra, the species observed corresponded with either kemptide or pkemptide protonated molecule ion [Mu+H]+ or [Mp+H]+ (calculated monoisotopic [Mu/p+H] +: = 772.5 Da or 852.5 Da respectively)
In order to test the sensitivity after electrophoresis, four different amounts of pkemptide (which included a trace amount of 32P pkemptide) ranging from 1 ng to 20 ng were spotted onto cellulose plates and submitted to electrophoresis under acidic conditions (pH 1.9). The samples were extracted as described and an aliquot corresponding to approximately 10% of the decant was withdrawn from each tube and analyzed with MALD. The positive MALD mass spectra of 1 and 5 ng samples of pkemptide spotted onto a cellulose plate and recovered using the procedure described showed species corresponding with the intact molecule ion. A number of other species were observed at the lowest concentration (1 ng) measured including the [Mp+Na]+ (m/z 874) and two fragment ions (m/z 749 and 772). The m/z 772 fragment ion is consistent with the mass of non phosphorylated kemptide and is therefore assigned as facile loss of HPO3 (80 Da) occurring in the source during the ionization process. In order to further investigate the nature of the m/z 749 fragment ion we measured the post source decay spectrum of 20 ng of pkemptide. An intense fragment ion is observed at the apparent m/z 749 of the uncalibrated FAST spectrum (i.e., prior to setting the command "FastSetMeth" to "UsedFragIons"). After setting the command "FastSetMeth" to "UsedFragIons", the measured mass of the m/z 749 fragment ion was adjusted to m/z 754 (5) . We therefore assign the m/z 749 fragment ion in Figure 5 as loss of H2PO4 (97 Da) from the [Mp+H]+ occurring as a post source decay fragmentation.

Discussion
Electrophoresis and chromatography on thin layer cellulose plates are common laboratory procedures for phosphopeptide mapping. Using this technique, an enzymatic map of a protein may be used to identify putative peptide fragments which are phosphorylated. The additional negative charge concomitant with phosphorylation serves to help identify the phosphopeptides on the 2D plates. However, even when the protein sequence is known, other techniques such as site directed mutagenesis or manual Edman chemistry must be used to determine the identity of the peptide. The procedure described here indicates that MALDI-TOF mass spectrometry may serve to identify peptides from such a proteolytic map. Previously, we and others have shown that sub pmol amounts of phosphorylated peptides can be analyzed with MALDI (6, 7) . As shown above MALDI is sensitive to 0.1 ng ("100 fmol) of kemptide. The MALDI spectrum of the 1 ng sample submitted to electrophoresis and extracted verifies this level of sensitivity. Although this spectrum contained some fragment ions in addition to the intact molecule ions, it is suggested that this is not a major limitation. This protocol is intended for identifying a peptide given a number of known possible proteolytic fragment sequences, where other ions may not interfere. The analysis of 1 ng of kemptide extracted without electrophoresis is consistent with this level of sensitivity (since approximately 10% of the extracted sample is actually used for the MS analysis). The electrophoresis step does not alter the overall sensitivity observed suggesting that the sample losses due to electrophoresis and extraction are minimal. In fact, the sensitivity is equivalent if not better than that observed for direct MALD-MS measurements from a thin layer plate (8) . We suggest that improved sensitivity may be afforded if a larger proportion of the extracted sample was analyzed. Currently we are investigating appropriate conditions for pre concentration of 10-100 ml volumes of nM concentration solutions.

Summary
MALDI TOF can conveniently be used to analyze sample directly extracted from thin layer cellulose plates after electrophoresis. The technique is sensitive to <5 ng of peptide applied to the cellulose plate.

Experimental Procedures
The MALDI mass spectra were measured using a 'Bruker Reflex' (Bruker Instruments Inc. Bremen, Germany) MALDI time-of-flight mass spectrometer equipped with a nitrogen laser (337 nm). Spectra were measured in both the linear and reflectron modes of operation. The accelerating potential of the Bruker instrument was +31 kV. The mass accuracy in the reflectron mode with external calibration of these instruments was typically better than 100 p.p.m (9) . The MALDI spectra represent the accumulation of 20- 70 laser shots, fired at the sample, which was located by revolving the sample probe. The sample matrix a-cyano-4-hydroxycinnamic acid (ACHC) (Aldrich Chemical Co., Milwaukee, Wis. Catalog No. 14550-5) (10) dissolved in acetone was prepared as described by Vorm et al. (11) . A 1.0 ml aliquot of the pkemptide was applied on the thin ACHC film (0.5 ml) which had previously been deposited on a stainless steel target. The solution was then left to dry at room temperature, rinsed with 5 ml de-ionized water, dried with a stream of nitrogen gas and inserted into the mass spectrometer and analyzed.

Phosphorylation
Kemptide was purchased (Sigma Chemical Co., St. Louis) and enzymatically phosphorylated in vitro with the catalytic subunit of PK-A (obtained from S. Taylor, UCSD) in the presence of ATP (or 32P ATP to generate tracer) in kinase buffer (25 mM Tris, 10 mM MgCl2, 2 mM EGTA, 1 mM DTT, pH 7) for 1 hour at room temperature. Phosphopeptides were then isolated by reversed phase HPLC.

Electrophoresis
Phosphopeptide samples (containing approx. 10,000 cpm 32P labeled peptide) were spotted onto cellulose plates (Merck, Catalog No. 5716) and electrophoresis carried out in pH 1.9 buffer at 500V for 5 min. as described in (12) . The plates were then dried in an air stream and autoradiographed until the location of the peptide could be identified. The spots were marked with a soft pencil and the cellulose powder was scraped with a pointed spatula and collected into a filter-containing micro centrifuge tip attached to a vacuum line. The peptide-containing powder was then transferred to a micro centrifuge tube by centrifugation.

Extraction
A 10 ml aliquot of a 0.1 % TFA and 25% acetonitrile solution was added to each micro centrifuge tube containing cellulose powder. The resulting suspension was floated in an ultrasonic bath for 5 minutes. The micro centrifuge tubes were then centrifuged for 1 minute prior to removing a 1 ml aliquot of the decant for MALDI analysis. Note: the use of smaller extraction volumes (e.g. 5 ml aliquots) were attempted, however, a substantial proportion of the solution added is adsorbed by the cellulose powder and the 10 ml aliquot used was the smallest volume which allowed for centrifugal separation of the sample.

REFERENCES
1. N. G. Ahn, R. Seger, R. L. Bratlien and E. G. Krebs (1992). Growth factor-stimulated phosphorylation cascades: activation of growth factor-stimulated MAP kinase. Ciba Found. Symp. 164, 113-126

2. J. Posada and J. A. Cooper (1992). Molecular signal integration. Interplay between serine, threonine, and tyrosine phosphorylation. Molecular Biology of the Cell 3, 583-592

3. T. Hunter and M. Karin (1992). The regulation of transcription by phosphorylation. Cell 70, 375-387

4. D. W. Meek and A. J. Street (1992). Nuclear protein phosphorylation and growth control. Biochem. J. 287, 1-5

5. G. Kruppa (May 3, 1994). Application Note: Obtaining FAST Spectra with the REFLEX

6. A. G. Craig, C. A. Hoeger, C. L. Miller, T. Goedken, J. E. Rivier and W. H. Fischer (1994). Monitoring Protein Kinase and Phosphatase Reactions with Matrix Assisted Laser Desorption/Ionization and Capillary Zone Electrophoresis: Comparison of the Detection Efficiency of Peptide-Phosphopeptide Mixtures. Biol. Mass Spectrom. 23, 519-528

7. Y. K. Wang, P. C. Liao, J. Allison, D. A. Gage, P. C. Andrews, D. M. Lubman, S. M. Hanash and J. R. Strahler (1993). Phorbol 12-Myristate 13-Acetate-Induced phosphorylation of op18 in jurkat T-Cells - identification of phosphorylation sites by Matrix-Assisted laser desorption ionization mass spectrometry. J Biol Chem 268, 14269-14277

8. A. I. Gusev, A. Proctor, Y. I. Rabinovich and D. M. Hercules (1995). Thin-layer chromatography combined with matrix-assisted laser desorption ionization mass spectrometry. Anal Chem 67, 1805-1814

9. O. Vorm and M. Mann (1994). Improved Mass Accuracy in Matrix Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry of Peptides. J Am Soc Mass Spectrom 5, 955-958

10. R. C. Beavis and B. T. Chait (1992). a-Cyano-4-hydroxycinnamic Acid as a Matrix for Matrix assisted Laser Desorption mass Spectrometry. Org Mass Spectrom 27, 156-158

11. O. Vorm, P. Roepstorff and M. Mann (1994). Improved resolution and very high sensitivity in MALDI TOF of matrix surfaces made by fast evaporation. Anal Chem 66, 3281-3287

12. W. J. Boyle, P. Van der Greer and T. Hunter (1991). Phosphopeptide Mapping and phosphoamino Acid Analysis by Two-Dimensional Separation on Thin-Layer Cellulose Plates. B. M. Sefton and T. Hunter (ed) In: Analysis of Protein Phosphorylation, Academic Press, San Diego. 110-149

13. A. G. Craig, S. C. Koerber, J. Porter, C. Hoeger and J. E. Rivier (1993). An expanded nomenclature scheme for labeling peptide fragmentations and its use with AMASS, a computer program for generating all possible fragment ion structures from known precursors. Biol. Mass Spectrom. 22, 31-44

Back to Home Page