On-Probe Sample Preparation without Washes for Matrix-Assisted

On-Probe Sample Preparation without Washes for Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Using an Anion Exchange Medium...
3 downloads 0 Views 247KB Size
Download PDF
Anal. Chem. 2005, 77, 5750-5754

Technical Notes

On-Probe Sample Preparation without Washes for Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Using an Anion Exchange Medium Seketsu Fukuzawa,†,‡ Miwako Asanuma,‡ Kazuo Tachibana,† and Hiroshi Hirota*,‡,§

Department of Chemistry, School of Science, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan, Protein Research Group, RIKEN Genomic Sciences Center, Suehiro-cho, Tsurumi-ku, Yokohama 230-0045, Japan, and Division of Supramolecular Biology, Yokohama City University, Suehiro-cho, Tsurumi-ku, Yokohama 230-0045, Japan

When the mass spectra of biological samples (proteins, peptides, and so on) are obtained routinely by matrixassisted laser desorption/ionization mass spectrometry (MALDI-MS), a serious problem is the reduction of the ionization efficiency by impurities, such as buffer salts and detergents. We focused our attention on devising a method to maintain the ionization efficiency of protein samples, even in the presence of sodium dodecyl sulfate (SDS), without any extra purification step. Although no protein ion peaks are observed in the presence of 2.5% SDS with the usual methods, the addition of a granular anion exchange silica gel to the matrix solution allowed the protein ion peaks to be obtained with an excellent signal-to-noise ratio. Together with other supporting experiments, we suggest that the positively charged surface (the basic environment derived from the anion exchange groups) and the roughness of the particles were important for good ionization in the presence of a high SDS concentration. For a very uneven surface, the SDS might be absorbed into the particle interiors during the process of cocrystallization with the matrix and analytes, which is known as the molecular sieve effect, and the SDS concentration in the surface crystalline film might be reduced. As a result, we developed an on-probe sample preparation method without washes for MALDI, using a strong anion exchange silica gel. This method is applicable even in the presence of 2.5% SDS, and is not only very simple but also inexpensive, because it can be used with the standard MALDI target plates. In the postgenome sequencing era, comprehensive analyses of various proteomes are being attempted, which require further improvements for the quick and accurate identification of pro* Corresponding author. Tel: +81-45-503-9211. Fax: E-mail: [email protected]. † The University of Tokyo. ‡ RIKEN Genomic Sciences Center. § Yokohama City University.

+81-45-503-9210.

5750 Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

teins.1,2 Proteome research using mass spectrometry (MS) has been supported by the revolutionary progress in soft ionization technology as applied to biopolymers, such as the development of electrospray ionization3 (ESI) and matrix-assisted laser desorption/ionization4,5 (MALDI) techniques.6-8 Two-dimensional polyacrylamide gel electrophoresis9,10 followed by MS is one of the mainstream analytical techniques in proteome research. One of the very serious problems for proteomics by MS (routinely obtaining huge numbers of mass spectra) is the lowering of the ionization efficiency by impurities, such as buffer salts and detergents. These additives are indispensable for handling protein samples, to maintain their activities and folding. However, these additives should be removed for MS identification of protein samples. In the case of ESI, this problem has been conquered by combining liquid chromatography with ESI (LC-ESI), and this combination is another mainstream technique in proteome analyses.11,12 In the case of MALDI, however, there are still several problems with high-throughput sample preparation techniques.13-15 (1) Celis, J. E.; Ratz, G. P.; Madsen, P.; Gesser, B.; Lauridsen, J. B.; Kwee, S.; Rasmussen, H. H.; Nielsen, H. V.; Cruger, D.; Basse, B.; Leffers, H.; Honore, B.; Moller, O.; Celis, A.; Vandekerckhove, J.; Bauw, G.; Vandamme, J.; Puype, M.; Vandenbulcke, M. FEBS Lett. 1989, 244, 247-254. (2) Lottspeich, F. Angew. Chem., Int. Ed. 1999, 38, 2477-2492. (3) Fenn, J. B.; Mann, M,; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science 1989, 246, 64-67. (4) Tanaka, K.; Waki, H.; Ido, Y.; Akita, S.; Yoshido, Y. Rapid Commun. Mass Spectrom. 1988, 2, 151-153. (5) Karas, M.; Hillenkamp, F. Anal. Chem. 1988, 60, 2299-2301. (6) Lee, T. D.; Shively, J. E. Curr. Opin. Biotechnol. 1991, 2, 52-60. (7) Roepstorff, P. Curr. Opin. Biotechnol. 1997, 8, 6-13. (8) Henzel, W. J.; Watanabe, C. J. Am. Soc. Mass Spectrom. 2003, 14, 931942. (9) O’Farrell, P. H. J. Biol. Chem. 1975, 250, 4007-4021. (10) Henzel, W. J.; Billeci, T. M.; Stults, J. T.; Wong, S. C.; Grimley, C.; Watanabe, C. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 5011-5015. (11) Vissers, J. P. C.; Chervet, J. P.; Salzmann, J. P. J. Mass Spectrom. 1996, 31, 1021-1027. (12) Bergquist, J.; Palmbladm, M.; Wetterhall, M.; Hakansson, P.; Markides, K. E. Mass Spectrom. Rev. 2002, 21, 2-15. (13) Ehring, H.; Stromberg, S.; Tjernberg, A.; Noren, B. Rapid Commun. Mass Spectrom. 1997, 11, 1867-1873. (14) Puchades, M.; Westman, A.; Blennow, K.; Davidsson, P. Rapid Commun. Mass Spectrom. 1999, 13, 344-349. (15) Xu, Y.; Bruening, M. L.; Watson, J. T. Mass Spectrom. Rev. 2003, 22, 429440. 10.1021/ac0500129 CCC: $30.25

© 2005 American Chemical Society Published on Web 07/29/2005

Before the matrix-analyte cocrystalline films are built on the target probe, a short chromatography step using reversed-phase media is usually employed, but not only is this an additional step it also requires some technical skill.16,17 One group reported the removal of impurities on probes coated with self-assembled monolayers.18,19 This method does not employ chromatography but is based on the concept of removing impurities before matrixanalyte cocrystallization. Recently, a promising alternative was reported by Rajnarayanan and Wang.20 They proposed an ion-pair assisted recovery technique, where cationic groups of ion pairing agents, such as long-chain alkylammonium salts, interact electrostatically with the anionic sulfate group of sodium dodecyl sulfate (SDS) or another anionic surfactant, and the alkyl chains of both molecules interact hydrophobically to form ion pairs that are neutral and amphiphilic. We also focused on a “simple pretreatment” process, rather than “removing” the impurities, and intended to develop this concept into an on-probe sample preparation method without washes for MALDI. Typical compounds used for the matrix with a nitrogen laser (wavelength, 337 nm) are 3,5-dimethoxy-4-hydroxycinnamic acid (sinapinic acid),21 R-cyano-4-hydroxycinnamic acid (R-CHCA),22 3-hydroxypicolinic acid (3-HPA),23 picolinic acid (PA),24 trans-3indoleacrylic acid,25 and 2,5-dihydroxybenzoic acid (gentisic acid),26 which are all acidic compounds. In terms of the ion pairing idea, we hypothesized that matrix-analyte cocrystalline films should be formed on an anion exchange medium surface efficiently by electrostatic interactions. Matrix-analyte cocrystalline films are usually developed in two dimensions on the probe; therefore, three-dimensional development of the matrix-analyte cocrystalline films on a surface of tiny particles will promise larger surface area. The increase of the irradiated surface area of the matrix-analyte cocrystalline films without increasing the laser focus would result in desorption of a larger amount of analyte molecule. Here we describe a simple sample preparation technique, using an anion exchange medium, for MALDI experiments carried out in the presence of buffer salts and SDS, which is one of the most powerful surfactants in common use. EXPERIMENTAL SECTION Materials. Cytochrome c (MW 12 360), myoglobin (MW 16 951), adrenocorticotropic hormone fragment 7-38 (ACTH (16) Erdjument-Bromage, H.; Lui, M.; Lacomis, L.; Grewal, A.; Annan, R. S.; McNulty, D. E.; Carr, S. A.; Tempst, P. J. Chromatogr., A 1998, 826, 167181. (17) Rouse, J. C.; Vath, J. E. Anal. Biochem. 1996, 238, 82-92. (18) Brockman, A. H.; Dodd, B. S.; Orland, R. Anal. Chem. 1997, 69, 47164720. (19) Warren, M. E.; Brockman, Orland, R. Anal. Chem. 1997, 70, 37573761. (20) Rajnarayanan, R. V.; Wang, K. J. Mass Spectrom. 2004, 39, 79-85. (21) Beavis, R. C.; Chait, B. T. Rapid Commun. Mass Spectrom. 1989, 3, 432435. (22) Beavis, R. C.; Chaudhary, T.; Chait, B. T. Org. Mass Spectrom. 1992, 27, 156-158. (23) Wu, K. J.; Steding, A.; Becker, C. H. Rapid Commun. Mass Spectrom. 1993, 7, 142-146. (24) Tang, K.; Taranenko, N. I.; Allman, S. L.; Chen, C. H.; Chang, L. Y.; Jacobson, K. B. Rapid Commun. Mass. Spectrom. 1994, 8, 673-677. (25) Chou, J. Z.; Kreek, M. J.; Chait, B. T. J. Am. Soc. Mass Spectrom. 1994, 5, 10-16. (26) Strupat, K.; Karas, M.; Hillenkamp, F. Int. J. Mass Spectrom. Ion Processes 1991, 111, 89-102.

7-38, MW 3659.2), insulin (MW 5733.5), ubiquitin (MW 8564.8), trypsin (MW 23 311), enolase (MW 46 670), and bovine serum albumin (BSA, MW 66 430) were obtained from Sigma (St. Louis, MO). Wako-gel LC-SAX-10H (10-µm quaternary ammonium functionalized granular silica gel) was supplied by Wako (Osaka, Japan). Partisil 10 SCX (10-µm sulfonated granular silica gel), Partisil 10 (10-µm granular silica gel), Nucleosil 100-SB10 (10-µm quaternary ammonium functionalized spherical silica gel), and Nucleosil 100-SB5 (5-µm quaternary ammonium functionalized spherical silica gel) were obtained from GL-Science (Tokyo, Japan). Toyopearl SuperQ-650S (the 20-50-µm quaternary ammonium functionalized spherical vinyl polymer) and Toyopearl DEAE-650S (20-50-µm diethylaminoethyl groups functionalized spherical vinyl polymer) were purchased from TOSOH (Tokyo, Japan). These commercial materials were used without further purification. Sinapinic acid was purchased from TCI (Tokyo, Japan) and was further purified by recrystallization from acetonitrile twice prior to use. All water used was obtained from a Milli-Q system (Millipore, Bedford, MA). Sample Preparation. The sinapinic acid matrix, with or without the solid-phase material suspension (50 mg/mL), was prepared from a stock solution of 10 mg/mL sinapinic acid in 50% aqueous acetonitrile with 0.1% trifluoroacetic acid. Protein sample solutions were prepared by combining equal volumes of a protein solution in a buffer containing 100 mM NaCl and 50 mM TrisHCl (pH 7.5) and a surfactant (SDS (MW 288.38), 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS, MW 614.9), polyoxyethylene(10) isooctylphenyl ether (Triton X-100, MWav 625), sodium deoxychlolate (DOC, MW 414.6), 1-O-octyl β-D-glucopyranoside (OG, MW 292.4), polyoxyethylene(4) lauryl ether (Brij 30, MWav 362), polyoxyethylene(20) sorbitan monolaurate (Tween 20, MWav 1228), polyoxyethylene(23) lauryl ether (Brij 35, MWav 1198), or cethyltrimethylammonium bromide (CTAB, MW 364.5)) solution. The final concentrations in the sample solution was 100 µM myoglobin, 10 µM cytochrome c, 8.3 µM ACTH 7-38, 8.3 µM insulin, 8.3 µM ubiquitin, 25 µM trypsin, 25 µM enolase, and 25 µM BSA, in a buffer of 50 mM NaCl and 25 mM Tris-HCl (pH 7.5), and 1, 1.5, 1.8, 2, 2.5, 3, 5, or 6% surfactant (w/v), respectively. Sample preparation involved the dried droplet method.27 A small aliquot (0.1 µL) of sample solution was spotted, and then 0.5 µL of the matrix suspension was added to each target well of an Applied Biosystems (Foster City, CA) Voyager DE STR 100well sample plate, pretreated to form the thin layer. White opaque matrix-analyte cocrystalline films began to form in the target well within 10 min at room temperature. MALDI-MS Experiments. MALDI-MS experiments were carried out with an Applied Biosystems Voyager DE STR mass spectrometer, using a nitrogen laser at 337 nm with an initial accelerating voltage of 25 kV, an extraction delay time of 500 ns, and laser intensities of 1700 (MW < 10 000) or 1900 (MW > 10 000). Linear mode operation was performed in the positive ion mode. All spectra were obtained from a sum of 100 acquisition shots. Mass calibration was performed on the single- and doublecharged ion peaks of myoglobin, insulin, or BSA, as external references. All experiments were carried out at least four times to ensure the reproducibility of the spectral profiles. (27) Beavis, R. C.; Chait, B. T. Methods Enzymol. 1996, 270, 519-551.

Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

5751

Figure 2. MALDI mass spectra of the mixture of ACTH 7-38, insulin, and ubiqutin in the presence of 2.5% SDS. Matrix-analyte cocrystallized with (A) and without (B) Wako-gel LC-SAX-10H. Protein symbols: 9, ACTH 7-38; 4, insulin; [, ubiquitin.

Figure 1. MALDI mass spectra of cytochrome c and myoglobin in the presence of 2.5% SDS. Matrix-analyte cocrystallized with no additive (A), silica gel (B), cation exchange silica gel (C), and anion exchange silica gel (D), respectively. Protein symbols: g, cytochrome c; b, myoglobin.

RESULTS AND DISCUSSION Effect of Solid-Phase Particles. Four experiments (a mixture of cytochrome c and myoglobin with no additive, with silica gel, with cation exchange silica gel, and with anion exchange silica gel) were carried out in the presence of 2.5% SDS. Although no protein ion was observed without additive (Figure 1A), a tiny, single-charged ion peak of cytochrome c appeared in the presence of the granular plain silica gel, Partisil 10 (Figure 1B). Both the single- and double-charged ion peaks of cytochrome c were more distinguishable after cocrystallization with the granular cation exchange silica gel, Partisil 10 SCX (Figure 1C). However, no ion peak could be reliably assigned to myoglobin in these three mass spectra. The use of a granular anion exchange silica gel (Wakogel LC-SAX-10H) as the additive afforded the prominent ion peaks of cytochrome c and myoglobin, with excellent signal-to-noise (S/ N) ratios (Figure 1D).28 Even though no sample ion peak could be observed at the higher concentrations of SDS (5% or more), a 2-fold dilution of the protein sample with water, to reduce the SDS concentration before matrix-analyte cocrystallization, was effective in the 5% SDS case (data not shown). Dilution of the protein samples reduces the concentration not only of SDS but also of proteins and buffer salts. Presumably, the reduction of the salt concentration assists in the ionization of proteins. Furthermore, the effect to use the granular anion exchange silica gel as the additive was confirmed in the ionization of a wide range of protein samples with different molecular weights (from 3 to 66 kDa) in the presence of 1% SDS (Figures 2 and 3). When preparing samples for MALDI, it is generally very important to minimize impurities, such as buffer salts and (28) Kobayashi, T.; Kawai, H.; Suzuki, T.; Kawanishi, T.; Hayakawa, T. Rapid Commun. Mass Spectrom. 2004, 18, 1156-1160.

5752 Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

Figure 3. MALDI mass spectra of BSA in the presence of 2.5% SDS. Matrix-analyte cocrystallized with (A) and without (B) Wakogel LC-SAX-10H. Protein symbol: ., BSA.

surfactants, when creating matrix-analyte cocrystals. In our experience, when the Wakogel LC-SAX-10H was removed from the matrix-analyte cocrystalline films following the procedure reported by Rouse and Vath17 after preparation under the same conditions used for Figure 1D, almost all of the signals originating from the protein samples were reduced to the noise level (data not shown). Effect of Functional Groups of Particles. Only the anion exchange silica gel enhanced the sample ionization in the presence of high concentrations of SDS, so we checked the differences in the functional groups on the solid phase (anion exchange silica gel) to explain these results. The Toyopearl SuperQ650S29 and Toyopearl DEAE-650S30 media have strong and weak anion exchange groups functionalized on the vinyl polymer, respectively.31 Toyopearl SuperQ-650S was a better medium than Toyopearl DEAE-650S for protein sample ionization in the presence of 1% SDS (Figure 4). The ionization efficiency of the sample might be affected by the amount of matrix-adsorbed medium. These experimental results can be explained in two ways. First, the matrix-analyte cocrystalline films, originating from the electrostatic interactions between the sinapinic acid and the quaternary ammonium groups, were formed on the (29) Staby, A.; Jensen, I. H. J. Chromatogr., A 2001, 908, 149-161. (30) Stocchi, V.; Masat, L.; Biagiarelli, B.; Accorsi, A.; Piccoli, G.; Palma, F.; Cucchiarini, L.; Dacha, M. Prepr. Biochem. 1992, 22, 11-40. (31) The reason we had to use this polymer instead of silica gel was that 10-µm diethylaminoethyl group functionalized granular silica gel was not commercially available.

Figure 4. MALDI mass spectra of cytochrome c and myoglobin in the presence of 1% SDS. Matrix-analyte cocrystallized with Toyopearl SuperQ-650S (A) and Toyopearl DEAE-650S (B). Protein symbols: g, cytochrome c; b, myoglobin.

anion exchange medium surface, and second, these immobilized ammonium groups stripped the SDS from the protein sample during the cocrystallization process and laser desorption.20 Effect of Particle Size and Shape. Next, the size and shape of the anion exchange particles were investigated. Although 10µm particles (Wakogel LC-SAX-10H) enabled us to observe the protein ion peaks with 2.5% SDS, this effect was not seen in the medium with the larger particle size (Toyopearl SuperQ-650S; 20-50 µm). This result reveals the importance of the surface area. That is, the smaller particles have larger surface areas. Furthermore, the granular-type particles have a much larger surface area than the spherical particles.32,33 Additional experiments were performed with the spherical anion exchange silica gels, Nucleosil 100-SB10 (10 µm) and Nucleosil 100-SB5 (5 µm). However, no ion peaks originating from cytochrome c and myoglobin were observed at all in the presence of 2.5% SDS. This suggests that both the positively charged surface and the particle shape are important for good ionization. For quite irregular surfaces, the low molecular weight buffer salts and surfactants, such as SDS, might be absorbed inside the particles during the process of cocrystallization with the matrix. This is known as the molecular sieve effect, and through this effect, the SDS concentration in the surface matrix-analyte cocrystalline films might be reduced.34 This concept is supported by the fact that the intensity of the cluster ion peaks around the m/z 1000-4000 of SDS became lower in the presence of Wakogel LC-SAX-10H (data not shown). Our initial working hypothesis was that the increase of the irradiated surface area of the matrix-analyte cocrystalline films without increasing the laser focus would result in desorption of larger amount of analyte molecules. However, the intensity of protein sample ion peaks was reduced in the presence of the Wakogel LC-SAX-10H at buffer salt and surfactant-free condition (data not shown). This implies our working hypothesis is not correct and the anion exchange media might play an important role for ionization efficiency. (32) Fieser, L. F.; Fieser, M. Reagents for Organic Synthesis; John Wiley and Sons: New York, 1967; pp 703-705. (33) Beck, J. S.; Vartuli, J. C. Curr. Opin. Solid State Mater. 1996, 1, 7687. (34) Kisler, J. K.; Da¨hler, A.; Stevens, G. W.; O’Connor, A. J. Microporous Mesoporous Mater. 2001, 44-45, 769-774.

Figure 5. MALDI mass spectra of cytochrome c and myoglobin in the presence of 3% of CHAPS (A) and Triton X-100 (B). Matrixanalyte cocrystallized with (left) or without (right) Wako-gel LC-SAX10H. Protein symbols: g, cytochrome c; b, myoglobin.

Effect for Other Surfactants. We have emphasized that our method is effective for SDS-containing samples, but we also confirmed the utility of this method with other low molecular weight surfactants, such as CHAPS, Triton X-100 (Figure 5) and DOC, OG, and Brij 30 (data not shown). These surfactants are also commonly used in protein experiments (expression, purification, and so on). However, our method is not effective for ionization of protein samples in the presence of high molecular weight surfactants (Tween 20, Brij 35) and a cationic surfactant (CTAB) (data not shown). For protein identification, our simple and quick sample preparation method for MALDI-MS can be an alternative to SDS-PAGE. CONCLUSION In the presence of high concentration of SDS, 10-µm quaternary ammonium functionalized granular silica gel is the best matrix-analyte cocrystallization medium for protein sample ionization. The basic environment derived from strong anion exchange groups (quaternary ammonium derivatives) and the surface unevenness that occurs with granular-type particles are two of the most significant factors for maintaining the high ionization efficiency of protein samples in the presence of a high concentration (2.5%) of SDS. In the basic environment, the matrix-analyte cocrystalline film surface consisting of sinapinic acid (matrix) and the protein sample can be efficiently developed. The surface irregularities might help to trap low molecular weight impurities, such as surfactants or buffer salts, inside the granular particles by the molecular sieve effect, in combination with ionpair assisted recovery.20 A MALDI target plate consists of hundreds of target wells, and some mass spectrometers have multiple target plate stockers for high-throughput screening. The method described here involves Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

5753

on-probe sample preparation without washes. Although this method is derived from the classical dried droplet method upon a normal stainless plate, we can handle proteins with no loss of sample content and observe the protein sample ion peaks even in the presence of a high concentration of SDS. The ease and low expense of this technique provide further advantages. Improvements in instrumentation, combined with advances in methodology, such as those described here, will enable the analyses of the full length proteins by MS/MS measurements without proteolytic digestion soon.35,36 (35) Van Berkel, G. J. Eur. J. Mass Spectrom. 2003, 9, 539-562. (36) Cristoni, S.; Bernardi, L. R. Mass Spectrom. Rev. 2003, 22, 369-406.

5754

Analytical Chemistry, Vol. 77, No. 17, September 1, 2005

ACKNOWLEDGMENT Part of this work was supported by a Grant-in-Aid for Scientific Research on Priority Areas (C) Genome Science (I-Ab2-0300) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. SUPPORTING INFORMATION AVAILABLE Additional extensive figures. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review January 4, 2005. Accepted June 24, 2005. AC0500129