Enrichment and Separation of Mono- and Multiply Phosphorylated Peptides Using Sequential Elution from IMAC Prior to Mass Spectrometric Analysis

Transcription

1 Chapter 6 Enrichment and Separation of Mono- and Multiply Phosphorylated Peptides Using Sequential Elution from IMAC Prior to Mass Spectrometric Analysis Tine E. Thingholm, Ole N. Jensen, and Martin R. Larsen Summary Phospho-proteomics relies on methods for efficient purification and sequencing of phosphopeptides from highly complex biological systems using low amounts of starting material. Current methods for phosphopeptide enrichment, e.g., immobilized metal affinity chromatography and titanium dioxide chromatography, provide varying degrees of selectivity and specificity for phosphopeptide enrichment. Furthermore, the number of multiply phosphorylated peptides that are identified in most published studies is rather low. Here the protocol for a new strategy that separates mono-phosphorylated peptides from multiply phosphorylated peptides using sequential elution from immobilized metal affinity chromatography is described. The two separate phosphopeptide fractions are subsequently analyzed by mass spectrometric methods optimized for mono-phosphorylated and multiply phosphorylated peptides, respectively, resulting in improved identification of especially multiply phosphorylated peptides from a minimum amount of starting material. The new method increases the coverage of the phosphoproteome significantly. Key words: Phosphopeptide enrichment, Multiply phosphorylated peptides, Immobilized metal affinity chromatography (IMAC), Sequential elution, SIMAC, Titanium dioxide (TiO 2 ) chromatography, Mass spectrometry. 1. Introduction Several techniques exist for phosphopeptide enrichment prior to mass spectrometric analysis. Today the most commonly used methods are immobilized metal affinity chromatography (IMAC) and titanium dioxide (TiO 2 ) chromatography (see Chapters Marjo de Graauw (ed.), Phospho-Proteomics, Methods and Protocols, vol Humana Press, a part of Springer Science + Business Media, New York, NY Book DOI: / _6 67

2 68 Thingholm, Jensen, and Larsen Enrichment and Characterization of Phosphopeptides by Immobilized Metal Affinity Chromatography (IMAC) and Mass Spectrometry and The Use of Titanium Dioxide Micro-Columns to Selectively Isolate Phosphopeptides from Proteolytic Digests ). Recent studies comparing three different phosphopeptide enrichment methods, including phosphoramidate chemistry (PAC) (1), IMAC, and TiO 2 chromatography, showed that each method isolated distinct, partially overlapping segments of a phosphoproteome, whereas none of the tested methods were able to provide a whole phosphoproteome (2). One of the challenges in large-scale phospho-proteomics is the analysis of multiply phosphorylated peptides. Multiply phosphorylated peptides are in general suppressed in the ionization process in the mass spectrometric (MS) analysis in the presence of mono- or non-phosphorylated peptides and thereby the chance to detect and identify them decreases. In addition, most mass spectrometers are only able to perform a limited number of tandem MS experiments (MS/MS) in a given time period resulting in the negligence of the less abundant multiply phosphorylated peptides. Furthermore, collision-induced dissociation (CID) fragmentation of phosphopeptides usually results in the loss of phosphoric acid with poor peptide backbone fragmentation and consequently little sequence information is obtained. Phosphorylation-directed multistage tandem MS (pdms 3 ), where an ion originating from a neutral loss signal detected in the MS 2 is selected for a second round of CID (MS 3 ), provides better sequence information for mono-phosphorylated peptide ions with relatively high abundance. However, multiply phosphorylated peptides will result in additional losses of phosphoric acid. Optimized pdms 3 (3), multistage activation (MSA) (4), or electron capture/transfer dissociation (ECD/ETD) (5, 6) will most likely give better results for multiply phosphorylated peptides. Though, multiply phosphorylated peptides need to be separated from mono-phosphorylated peptides in order to analyze them using different mass spectrometric setups. Recently, we developed a new method for the separation of mono-phosphorylated peptides from multiply phosphorylated peptides where we are using sequential elution from IMAC (SIMAC) (3). In this strategy the peptide mixture is incubated with IMAC beads, which have a stronger selectivity for multiply phosphorylated peptides than for mono-phosphorylated peptides (7). After incubation, the sample is split in three elution fractions (see Fig. 1), an IMAC flow-through fraction, an acidic (1% TFA) fraction, and a basic (ph 11.3) fraction. The IMAC flow-through and acidic fractions that contain mainly mono-phosphorylated peptides in addition to a significant number of nonphosphorylated peptides are further enriched by

3 Enrichment and Separation of Mono- and Multiply Phosphorylated Peptides 69 Fig. 1. The setup used for the enrichment and separation of mono- from multiply phosphorylate peptides using the SIMAC strategy. The peptide sample is mixed with the IMAC beads and incubated for 30 min in a Thermomixer at room temperature. After incubation the beads are packed into a GELoader tip forming an IMAC micro-column. The IMAC flow-through is collected and further enriched using TiO 2 chromatography. The monophosphorylated peptides are eluted from the IMAC micro-column using acidic elution conditions (1% TFA, ph 1.0) and for complex samples this eluate is also further enriched using TiO 2 chromatography. The multiply phosphorylated peptides are subsequently eluted from the IMAC micro-column using basic elution conditions (ammonia water, ph 11.3). All three fractions are then ready for mass spectrometric analysis. TiO 2 chromatography to achieve pure mono-phosphorylated peptide fractions prior to tandem MS analysis. The basic fraction is analyzed directly by tandem MS analysis without further TiO 2 purification, as multiply phosphorylated peptides binds very hard to TiO 2 material and may be lost at this step (8). Furthermore, this fraction in general has a relatively low level of nonphosphorylated peptides, and thus no further enrichment is necessary. SIMAC greatly improves the number of phosphorylation sites identified even from very low amounts of starting material and offers a way to identify and characterize multiply phosphorylated peptides at large-scale levels (3).

4 70 Thingholm, Jensen, and Larsen 2. Materials 2.1. Model Proteins 2.2. Reduction, Alkylation, and Digestion of Model Proteins 2.3. Cell Culture and Lysis of Human Mesenchymal Stem Cells (hmscs) 2.4. Extraction, Reduction, Alkylation, and Digestion of hmsc Proteins 1. Transferrin (human). 2. Serum albumin (bovine). 3. β-lactoglobulin (bovine). 4. Carbonic anhydrase (bovine). 5. β-casein (bovine). 6. α-casein (bovine). 7. Ovalbumin (chicken). 8. Ribonuclease B (bovine pancreas). 9. Alcohol dehydrogenase (Baker s yeast). 10. Myoglobin (whale skeletal muscle). 11. Lysozyme (chicken). 12. α-amylase (Bacillus species). All proteins were from Sigma (St. Louis. MO). 1. Ammonium bicarbonate. 2. Dithiotreitol (DTT). 3. Iodoacetamide. 4. Modified trypsin. (see Note 1) 1. MEM (EARLES) media w/o phenol red, with glutamax-i (GibcoTM) containing 1% penicillin/streptomycin (GibcoTM) and 10% fetal bovine serum (FBS) (GibcoTM). 2. Phosphate saline buffer (PBS). 3. Phosphatase inhibitor cocktail 1 and 2 (P2850 and P5726, Sigma ). 4. Cell dissociation buffer (CDB) (GibcoTM). 1. Lysis buffer: 7 M urea, 2 M thiourea, 1% N-octyl glycoside, 40 mm Tris-base, 300 U benzonase. 2. Dithiotreitol (DTT). 3. Iodoacetamide. 4. Acetone. 5. Endoproteinase Lys-C (lysyl endopeptidase, WAKO). 6. Ammonium bicarbonate. 7. Modified trypsin (see Note 1).

5 Enrichment and Separation of Mono- and Multiply Phosphorylated Peptides Immobilized Metal Affinity Chromatography (IMAC) 2.6. Titanium Dioxide (TiO 2 ) Chromatography 2.7. Reversed Phase (RP) Chromatography 1. Ironcoated PHOS-select metal chelate beads (Sigma ). 2. IMAC Loading buffer: 0.1% trifluoroacetic acid (TFA), 50% acetonitrile, HPLC Grade. 3. GELoader tips from Alpha Laboratories (Hampshire, S050 4NU, UK). 4. IMAC Elution buffer 1: 1.0% TFA, 20% acetonitrile. 5. IMAC Elution buffer 2: 20 μl ammonia solution (25%), 980 μl UHQ water (ph 11.3). 6. Formic acid. 7. Water was obtained from a Milli-Q system (Millipore, Bedford, MA) (UHQ water). (see Notes 1 and 2). 1. Titanium dioxide (TiO 2 ) beads (Titansphere, 5 μm, GL sciences Inc., 2. GELoader tips (Eppendorf, Hamburg, Germany) or p10 pipette tips depending on the size of the column needed (see Note 3). 3. 3M Empore C8 disk (3M, Bioanalytical Technologies, St. Paul, MN). 4. Acetonitrile. 5. Syringe for HPLC loading (P/N , N25/500-LC PKT 5, SGE, Ringwood, Victoria, Australia) to create a small plug of C 8 membrane material. 6. TiO 2 loading buffer: 1 M glycolic acid in 5% TFA, 80% acetonitrile. 7. TiO 2 washing buffer: 1% TFA, 80% acetonitrile. 8. TiO 2 elution buffer 1: 20 μl ammonia solution (25%), 980 μl UHQ water. 9. TiO 2 elution buffer 2: 30% acetonitrile. 10. Formic acid. 1. POROS Oligo R3 reversed phase material (Applied Biosystems, Foster City, CA). 2. GELoader tips (Eppendorf, Hamburg, Germany) or p10 pipette tips depending on the size of the column needed (see Note 4). 3. 3M Empore C18 disk (3M, Bioanalytical Technologies, St. Paul, MN). 4. Syringe for HPLC loading (P/N , N25/500-LC PKT 5, SGE, Ringwood, Victoria, Australia) to create a small plug of C 18 membrane material.

6 72 Thingholm, Jensen, and Larsen 5. RP Loading buffer: 0.1% TFA. 6. RP Elution buffer (for LC-ESI MSMS analysis): 70% acetonitrile, 0.1% TFA. 7. DHB Elution buffer (for MALDI MS analysis): 20 mg/ml DHB in 50% acetonitrile, 1% ortho-phosphoric acid (Bie & Berntsen A/S, UK). 3. Methods The principle of the SIMAC method is illustrated in this chapter using a peptide mixture originating from tryptic digestions of 12 standard proteins (Model proteins) (see Notes 5 and 6). The protocol is then applied to enrich for phosphorylated peptides from whole cell lysates from 120 μg protein from human mesenchymal stem cells (hmscs). The SIMAC purification method is a simple and very straightforward method that can be adapted to already existing methods. It is fast and efficient for enrichment of phosphopeptides even from highly complex samples (3, 8). The experimental setup of the method is illustrated in Fig. 1. For illustrating the anticipated results, a peptide mixture originating from tryptic digestions of 12 standard proteins was prepared Model Proteins and Peptide Mixture 3.2. Human Mesenchymal Stem Cell (hmsc) Proteins See details in Subheading 3.1, Chapter Enrichment and Characterization of Phosphopeptides by Immobilized Metal Affinity Chromatography (IMAC) and Mass Spectrometry. The large-scale experiments shown in this article were performed using a peptide mixture originating from tryptic digestion of 120-μg total protein from cell extract from hmscs. 1. Grow human mesenchymal stem cells (hmsc-tert20) in T75 flasks in MEM media without phenol red, with glutamax- I containing penicillin/streptomycin and fetal bovine serum at 37 C until they reached 90% confluence. Wash the confluent cells once with PBS buffer (37 C) and add 5-mL media (37 C) to cover the cells. Add phosphatase inhibitor cocktails (50 μl of each) to the media. Incubate the cells with the phosphatase inhibitors for 30 min at 37 C. Wash the cells with ice-cold PBS buffer harvest them using cell dissociation buffer. 2. After harvesting the cells, resuspend the cell pellet in 1.5-mL lysis buffer. Sonicate the cells at 40% output with intervals three times 15 s on ice to disrupt the cells and incubate at 80 C for 30 min. Add dithiotreitol (DTT, 20 mm) and incubate the sample at room temperature for 35 min. After incubation, add

7 Enrichment and Separation of Mono- and Multiply Phosphorylated Peptides mm iodoacetamide and incubate for 35 min at room temperature in the dark. 3. Add 14-mL ice-cold acetone to the solution and incubate at 20 C for 20 min to precipitate the proteins. Pellet the proteins by centrifugation at 6,000 g for 10 min at 4 C. Lyophilize the pellet and store it at 20 C until further use. 4. Redissolve a total of 1 mg of the precipitated and lyophilized protein from hmsc (weight out) in 6 M urea, 2 M thiourea to a concentration of 50 μg protein/μl. Incubate the proteins with endoproteinase Lys-C (1 2% w/w) at room temperature for 3 h. After incubation, dilute the endoproteinase Lys-C digested sample five times with 50 mm NH 4 HCO 3 and add 1 2% (w/w) chemically modified trypsin. Incubate the sample at room temperature for 18 h Batch Incubation with IMAC Beads 3.4. Packing the IMAC Micro-Column 1. Always adjust the amount of IMAC beads to the amount of sample in order to reduce the level of nonspecific binding from nonphosphorylated peptides. For 1 pmol tryptic digest use 7-μL IMAC beads (see Chapter Enrichment and Characterization of Phosphopeptides by Immobilized Metal Affinity Chromatography (IMAC) and Mass Spectrometry ). For more complex samples, where more material is available, use more IMAC beads. The following is describing the protocol when using 120-μg tryptic digest. 2. Transfer 40 μl of IMAC beads to a fresh 1-mL Eppedorf tube. 3. Wash the IMAC beads twice using 200 μl of IMAC loading buffer. 4. Resuspend the beads in 150 μl of IMAC loading buffer and add the sample (here 12 μl). 5. Incubate the sample with IMAC beads in a Thermomixer (Eppendorf) for 30 min at 20 C. 1. Squeeze the end of a 200-μL GELoader tip to prevent the IMAC beads from leaking. 2. After incubation pack the beads in the constricted end of the GELoader tip by application of air pressure forming an IMAC micro-column essentially as described in Chapter Enrichment and Characterization of Phosphopeptides by Immobilized Metal Affinity Chromatography (IMAC) and Mass Spectrometry (9). 3. It is critical to collect the IMAC flow-through in a fresh 1.5- ml Eppendorf tube for further analysis by TiO 2 chromatography. 4. Wash the IMAC column using 50-μL IMAC loading buffer and pool this with the IMAC flow-through.

8 74 Thingholm, Jensen, and Larsen 5. Lyophilize the IMAC flow-through and wash and enrich for mono-phosphorylated peptides using TiO 2 chromatography (see Subheading ) Elution of Monophosphorylated Peptides from the IMAC Micro-Column 1. Elute the mono-phosphorylated peptides bound to the IMAC micro-column using 50 μl of IMAC elution buffer 1 (Fig. 1). 2. Dry the eluate by lyophilisation. 3. Redissolve the peptides in 5-μL 1% SDS and purify further by TiO 2 chromatography (see Subheadings ) Elution of Multiply phosphorylated Peptides from the IMAC Micro-Column 1. Elute the multiply phosphorylated peptides bound to the IMAC micro-column using 70 μl of IMAC elution buffer 2 (Fig. 1). 2. For LC-ESI MS n analysis this IMAC eluate should be acidified using 100% formic acid (ph should be 2 3) and desalted/ concentrated using reversed phase chromatography Packing the TiO 2 Micro-Column 1. Prepare six TiO 2 micro-columns of 4 mm by packing the TiO 2 beads on top of the C 8 membrane disks in P10 tips as described in Subheading 3.2, Chapter The Use of Titanium Dioxide Micro-Columns to Selectively Isolate Phosphopeptides from Proteolytic Digests Loading the Sample onto the TiO 2 Micro-Column 1. Divide the IMAC flow-through + wash between three different TiO 2 micro-columns. Do the same for the IMAC eluate. See details for loading onto the TiO 2 micro-columns in Subheading 3.3, Chapter SILAC for Global Phosphoproteomic Analysis Elution of Phosphorylated Peptides from the TiO 2 Micro-Column 1. Elute the phosphopeptides bound to each TiO 2 micro-columns using at least 35 μl of TiO 2 elution buffer 1 followed by an elution step using 1 μl of TiO 2 elution buffer 2 and pool the eluates as described in Subheading 3.4, Chapter The Use of Titanium Dioxide Micro-Columns to Selectively Isolate Phosphopeptides from Proteolytic Digests (see Notes 7 and 8) Packing a Reversed Phase (RP) Micro-Column for Desalting/Concentrating the Sample 1. Prepare a RP micro-column of ~5 6 mm by packing the R3 beads on top of the C 18 membrane disk in P10 tips as described for TiO 2 micro-columns in Subheading 3.2, Chapter The Use of Titanium Dioxide Micro-Columns to Selectively Isolate Phosphopeptides from Proteolytic Digests..

9 Enrichment and Separation of Mono- and Multiply Phosphorylated Peptides Loading the Enriched Phosphopeptides onto the RP Micro-Column Elution of Phosphorylated Peptides from the RP Micro- Column 1. Load the acidified sample slowly onto the RP micro-column (~1 drop/s). 2. Wash the RP micro-column using 30 µl of 0.1% TFA. 1. Elute the phosphopeptides from the RP micro-column using 70 μl RP elution buffer, followed by lyophilization of the phosphopeptides. (N.B. For simple samples, the peptides can be eluted off the RP micro-column directly onto the MALDI target using 1 μl DHB solution). 2. LC-ESI-MS n analysis: Redissolve the dried phosphopeptides in 0.5 μl 100% formic acid and immediately dilute to 10 μl with UHQ water. LC-ESI-MS n analysis of multiply phosphorylated peptides is improved by redissolving the phosphopeptides by sonication in an EDTA containing buffer prior to LC-ESI-MS n analysis (10). Fig. 2. Results obtained from 1 pmol peptide mixture using the SIMAC strategy. MALDI MS spectra of peptides identified from the IMAC flow-through after further enrichment using TiO 2 chromatography (a), MALDI MS spectra of peptides eluted from the IMAC micro-column using 1% TFA (b) and MALDI MS spectra of peptides eluted from the IMAC microcolumn using ammonia water (ph 11.30) (c). The number of phosphate groups on the individual phosphopeptides are indicated by #P. Asterisk indicates the loss of phosphoric acid. (see Note 9).

10 76 Thingholm, Jensen, and Larsen Figure 2 illustrates the results obtained by using the SIMAC strategy on 1 pmol standard peptide mixture. MALDI MS spectra of the IMAC flow-through after TiO 2 enrichment (a) and MALDI MS spectra of the phosphopeptides eluted from the IMAC material using acidic conditions (1% TFA, ph 1.0) (b) show that mainly mono-phosphorylated peptides are identified in these two peptide fractions. Mainly multiply phosphorylated peptides are identified when using basic elution conditions (ammonia water, ph 11.3) (c) (see Note 9). Figure 3 compares the number of mono-phosphorylated and multiply phosphorylated as well as nonphosphorylated peptides identified from 120-µg hmsc tryptic digests using either TiO 2 enrichment or the SIMAC strategy. The results show that mainly mono-phosphorylated peptides are identified by TiO 2 chromatography, whereas only 17% (54 of 325) of the identified peptides are multiply phosphorylated. In the SIMAC experiment, the IMAC flow-through as well as the acidic elution fraction consist mainly of mono-phosphorylated peptides and the level of multiply phosphorylated peptides identified is even lower than for TiO 2 chromatography (3% and 4%, respectively). However, the level of multiply phosphorylated peptides identified is significantly increased by basic elution (68%). In total SIMAC gives rise to more than three times as many multiply phosphorylated peptides identified from 120-µg hmsc tryptic digest when compared with TiO 2 chromatography. Fig. 3. Results obtained from the analysis of phosphorylated peptides from acetone precipitated proteins from human mesenchymal stem cells (hmscs) using TiO 2 chromatography or SIMAC. The number of mono-phosphorylated peptides, multiply phosphorylated peptides and nonphosphorylated peptides identified from 120-μg hmsc tryptic digests using TiO 2 enrichment and the number of mono-phosphorylated peptides, multiply phosphorylated peptides and nonphosphorylated peptides identified in the three peptide fractions obtained from SIMAC analysis of 120-μg hmsc tryptic digests.

11 Enrichment and Separation of Mono- and Multiply Phosphorylated Peptides Notes 1. It is important to obtain the highest purity of all chemicals used. 2. All solutions should be prepared in UHQ water. 3. The capacity of the TiO 2 beads have been tested; however, more experiments need to be performed. The capacity of the beads appears to be very high, but when working with high amounts of material it is recommended to use P10 pipette tips instead of GELoader tips and to split the peptide sample and load it onto several TiO 2 micro-columns. The resulting eluates can then be pooled prior to MS analysis. For the experiments using 120-μg hmsc protein, six TiO 2 micro-columns of ~4 mm were packed in P10 tips. 4. For the experiments using 120-μg hmsc protein, a R3 microcolumn of ~5 6 mm were packed in P10 tips. 5. Always start by testing the method using a model peptide mixture. It is important to freshly prepare the peptide mixture as peptides bind to the surface of the plastic tubes in which they are stored. In addition, avoid transferring the peptide sample between different tubes to minimize adsorptive losses of the sample. 6. The peptide mixture used for the experiment illustrated in this chapter contained peptides originating from tryptic digestions of 1 pmol of each of the 12 proteins. 7. Use at least 35 μl of TiO 2 elution buffer 1 due to the size of the TiO 2 micro-columns. It is important to follow the first elution step using 1 μl of TiO 2 elution buffer 2, to elute peptides bound to the C 8 membrane plug. 8. At this step, pool the eluates from the three TiO 2 micro-columns with IMAC flow-through + wash by eluting the phosphopeptides into the same Eppendorf tube. Do the same for the three TiO 2 micro-columns with IMAC eluates. 9. The results obtained using this protocol will differ according to the mass spectrometer used for the analysis of the phosphopeptides, not only between MALDI MS and ESI MS but also within different MALDI MS instruments, depending on laser optics, laser frequency, instrumental configuration, sensitivity etc. Acknowledgements This work was supported by the Danish Natural Science Research Council (grant no (M.R.L) ) and the Danish Strategic Research Council (O.N.J).

12 78 Thingholm, Jensen, and Larsen References 1. Zhou, H. L., Watts, J. D., and Aebersold, R. A. (2001) A systematic approach to the analysis of protein phosphorylation. Nat. Biotechnol. 19, Bodenmiller, B., Mueller, L. N., Mueller, M., Domon, B., and Aebersold, R. (2007) Reproducible isolation of distinct, overlapping segments of the phosphoproteome. Nat. Methods 4, Thingholm, T. E., Jensen, O. N., Robinson, P. J., and Larsen, M. R. (2007) SIMAC A phosphoproteomic strategy for the rapid separation of mono-phosphorylated from multiply phosphorylated peptides. Published online by MCP on November 27th 2007 (M MCP200). 4. Schroeder, M. J., Shabanowitz, J., Schwartz, J. C., Hunt, D. F., and Coon, J. J. (2004) A neutral loss activation method for improved phosphopeptide sequence analysis by quadrupole ion trap mass spectrometry. Analytical Chemistry 76, Chalmers, M. J., Hakansson, K., Johnson, R., Smith, R., Shen, J., Emmett, M. R., and Marshall, A. G. (2004) Protein kinase A phosphorylation characterized by tandem Fourier transform ion cyclotron resonance mass spectrometry. Proteomics 4, Schroeder, M. J., Webb, D. J., Shabanowitz, J., Horwitz, A. F., and Hunt, D. F. (2005) Methods for the detection of paxillin posttranslational modifications and interacting proteins by mass spectrometry. J Proteome Res 4, Ficarro, S. B., McCleland, M. L., Stukenberg, P. T., Burke, D. J., Ross, M. M., Shabanowitz, J., Hunt, D. F., and White, F. M. (2002) Phosphoproteome analysis by mass spectrometry and its application to Saccharomyces cerevisiae. Nat Biotechnol 20, Jensen, S. S., and Larsen, M. R. (2007) Evaluation of the impact of some experimental procedures on different phosphopeptide enrichment techniques. Rapid Commun. Mass Spectrom. 21, Gobom, J., Nordhoff, E., Mirgorodskaya, E., Ekman, R., and Roepstorff, P. (1999) Sample purification and preparation technique based on nano-scale reversed-phase columns for the sensitive analysis of complex peptide mixtures by matrix-assisted laser desorption/ionization mass spectrometry. J. Mass Spectrom. 34, Liu, S., Zhang, C., Campbell, J. L., Zhang, H., Yeung, K. K., Han, V. K., and Lajoie, G. A. (2005) Formation of phosphopeptidemetal ion complexes in liquid chromatography/electrospray mass spectrometry and their influence on phosphopeptide detection. Rapid Commun Mass Spectrom. 19,