Recovery of Gel-Separated Proteins for In-Solution Digestion and

Anal. Chem. 2001, 73, 5370-5377

Recovery of Gel-Separated Proteins for In-Solution Digestion and Mass Spectrometry Andreas P. Jonsson,† Youssef Aissouni,‡,§ Carina Palmberg,† Piergiorgio Percipalle,‡ Erik Nordling,† Bertil Daneholt,‡ Hans Jo 1 rnvall,† and Tomas Bergman*,†

Department of Medical Biochemistry and Biophysics and Department of Cell and Molecular Biology, Medical Nobel Institute, Karolinska Institutet, SE-171 77 Stockholm, Sweden.

A protocol for mass spectrometry of gel-separated proteins resulting in significantly increased sequence coverage and in improved possibilities for detection and identification of posttranslational modifications was developed. In relation to the standard in-gel digestion procedure, the sequence coverage using a combination of matrix-assisted laser desorption/ionization and electrospray ionization mass spectrometry was on the average increased by 30%. The method involves electroblotting of the gel-separated proteins to a poly(vinylidene difluoride) membrane. The proteins are extracted from the membrane using a solution of 1% trifluoroacetic acid in 70% acetonitrile and lyophilized. After reconstitution of the protein extract in digestion buffer, proteolytic cleavage is carried out insolution as opposed to the standard in-gel digestion procedure. This allows recovery of large and hydrophobic peptides for mass spectrometry and reduces the risk for entrapment of proteolytic peptides in the gel matrix. The method was applied to proteins in the 30-40-kDa range with highly different structural properties. The improved ability to localize and determine protein modifications is shown for N-terminal acetylation and methylation of a histidine residue. Furthermore, the method enables fast screening of homologous protein sequences. Identification of proteins and characterization of posttranslational modifications is becoming increasingly important as a consequence of the rapid pace at which genomic sequences are currently determined. Polyacrylamide gel electrophoresis, particularly in two dimensions, combined with mass spectrometry is a major route to correlation of genomic data with protein expression.1,2 Mass spectrometrical analysis of gel-separated proteins is mainly carried out on proteolytic peptides generated by tryptic digestion of the gel-immobilized protein.3 However, a complication * Corresponding author: (tel) +46-8-728 77 80; (fax) +46-8-33 74 62; (e-mail) [email protected]. † Department of Medical Biochemistry and Biophysics. ‡ Department of Cell and Molecular Biology, Medical Nobel Institute. § Present address: U119 INSERM, Laboratory of Experimental Cancerology, F-13009, Marseille, France. (1) Roepstorff, P. Curr. Opin. Biotechnol. 1997, 8, 6-13. (2) Pandey, A.; Mann, M. Nature 2000, 405, 837-846. (3) Hellman, U.; Wernstedt, C.; Gonez, J.; Heldin, C. H. Anal. Biochem. 1995, 224, 451-455.

5370 Analytical Chemistry, Vol. 73, No. 22, November 15, 2001

with the in-gel digestion procedure is that only a limited part of the protein sequence is normally recovered for analysis, frequently less than 50% when peptide mass mapping by matrix-assisted laser desorption/ionization (MALDI) mass spectrometry is used.4,5 This is mainly due to difficulties in extraction of large peptides from the gel matrix and to variable accessibility of the proteolytic enzyme to the protein. This situation is usually not satisfactory for location and determination of posttranslational modifications since important residues and protein segments are potentially excluded from analysis. Several approaches to isolate intact proteins from gel separations for analysis by mass spectrometry have been suggested. After staining, the protein spot or band can be excised and the protein eluted from the gel piece using electrophoresis6 or chemical extraction.7,8 However, the effect of nonspecific adsorption to gel matrix and surfaces can be substantial and losses significant at low-picomole levels. To increase the recovery of protein for MALDI mass spectrometry using passive elution from gels, cocrystallization with MALDI matrix added to the extraction solution has been reported.8 Generally, electroblotting to a membrane support is more efficient when a limited amount of gel-separated material is available. The SDS present in the preparations is a major problem preventing mass spectrometric analysis by protein adduct formation that deteriorates the quality of the spectra. Removal of SDS by treatment with organic solvents for subsequent analysis by MALDI mass spectrometry has been demonstrated, but only with small proteins and at fairly high picomole levels.9 Another approach using infrared laser irradiation to decompose the SDS/protein adducts during electrospray ionization (ESI) mass spectrometry of proteins in SDS solutions has also been suggested but, again, only tested with fairly small proteins of sizes less than 20 kDa.10 Gel-separated proteins are (4) Oppermann, M.; Cols, N.; Nyman, T.; Helin, J.; Saarinen, J.; Byman, I.; Toran, N.; Alaiya, A. A.; Bergman, T.; Kalkkinen, N.; Gonzalez-Duarte, R.; Jo¨rnvall, H. Eur. J. Biochem. 2000, 267, 4713-4719. (5) Alaiya, A. A.; Oppermann, M.; Langridge, J.; Roblick, U.; Egevad, L.; Bra¨ndstedt, S.; Hellstro ¨m, M.; Linder, S.; Bergman, T.; Jo ¨rnvall, H.; Auer, G. Cell. Mol. Life Sci. 2001, 58, 307-311. (6) Haebel, S.; Jensen, C.; Andersen, S. O.; Roepstorff, P. Protein Sci. 1995, 4, 394-404. (7) Ehring, H.; Stro ¨mberg, S.; Tjernberg, A.; Noren, B. Rapid Commun. Mass Spectrom. 1997, 11, 1867-1873. (8) Cohen, S. L.; Chait, B. T. Anal. Biochem. 1997, 247, 257-267. (9) Puchades, M.; Westman, A.; Blennow, K.; Davidsson, P. Rapid Commun. Mass Spectrom. 1999, 13, 344-349. (10) Fridriksson, E. K.; Baird, B.; McLafferty, F. W. J. Am. Soc. Mass Spectrom. 1999, 10, 453-455. 10.1021/ac010486h CCC: $20.00

© 2001 American Chemical Society Published on Web 10/06/2001

Figure 1. Schematic outline of the current protocol with recovery of intact proteins from gel separations for analysis by mass spectrometry. The one- or two-dimensionally separated protein is electrotransferred to a PVDF membrane, extracted from the membrane, and digested in-solution before desalting and analysis with MALDI and ESI mass spectrometry. The method provides a potentially complete peptide mass map of the protein forming a basis for efficient identification of posttranslational modifications via tandem mass spectrometry.

also electroblotted to membrane supports, e.g., poly(vinylidene difluoride) (PVDF), and stained with Coomassie blue or some other suitable dye for on-membrane digestion to generate fragments of the protein for further analysis.11 However, as for the in-gel digestion, the recovery of large peptides is rather low and the proteolytic enzyme does not have full access to the target protein since the latter is tightly immobilized onto a surface. Alternatively, the protein can be eluted from the membrane using detergent solutions, e.g., containing SDS and Triton X-100,12 but this route will generally increase adduct formation and suppression in subsequent mass spectrometry. In this report, we describe a protocol for improved recovery of protein fragments for analysis by mass spectrometry compared to use of the standard in-gel digestion procedure. The improved yield is because trypsin digestion is carried out in-solution after electroblotting to PVDF and extraction of the intact protein without the use of detergents (Figure 1). The proteolytic peptides are analyzed by MALDI and ESI mass spectrometry. The latter ionization technique is primarily used for tandem mass spectrometry experiments to determine the amino acid sequence and to explore posttranslational modifications. In addition, we show efficient use of the current protocol to analyze the structures of homologous proteins. EXPERIMENTAL SECTION Materials. The 40-kDa heterogeneous ribonuclear protein 36 (Hrp 36) and actin (42 kDa) were purified from the dipteran Chironomus tentans.13 Actin was also prepared from Bos taurus (calf thymus) as described.14 Bovine carbonic anhydrase (30 kDa), (11) Fernandez, J.; Andrews, L.; Mische, S. M. Anal. Biochem. 1994, 218, 112117. (12) Szewczyk, B.; Summers, D. F. Anal. Biochem. 1988, 168, 48-53. (13) Percipalle, P.; Zhao, J.; Pope, B.; Weeds, A.; Lindberg, U.; Daneholt, B. J. Cell Biol. 2001, 153, 229-235. (14) Vandekerckhove, J.; Weber, K. Eur. J. Biochem. 1978, 90, 451-462.

bovine serum albumin (67 kDa), and phosphorylase B (97 kDa) originated from a molecular weight standard mixture (Amersham Pharmacia Biotech). All chemicals were of analytical grade. SDS/Polyacrylamide Gel Electrophoresis and In-Gel Digestion. Proteins were electrophoresed according to Laemmli.15 Gloves were used at all times when the gels were handled. Precast 12% Tris-glycine gels of thickness 1 mm containing 12 wells (Novex) were used. Electrophoresis was carried out using an XCell II minicell apparatus (Novex) at room temperature. Protein samples were mixed 1:1 (v/v) with sample buffer (2× concentrated, Novex). The sample buffer contained β-mercaptoethanol as the reducing agent and bromophenol blue to visualize the electrophoresis front. The sample was briefly heated (85-90 °C, 2 min) before it was loaded onto the gel. The electrophoresis running buffer was prepared from a 10× concentrated stock solution (Novex) containing Tris base, glycine, and SDS. Electrophoresis was carried out at 125 V for ∼2 h (until the dye marker had reached the edge of the gel). After electrophoresis, proteins were either stained with Coomassie blue for subsequent in-gel digestion or they were electroblotted for extraction and in-solution digestion (below). For gel-staining, Coomassie R-250 was used at 0.1% (w/v) in 40% methanol containing 10% acetic acid for ∼4 h (shaking at room temperature). Destaining was carried out in the same methanol/acetic acid solution, but without Coomassie stain, until the protein bands were clearly visible against the background (usually overnight with shaking at room temperature). For in-gel digestion with trypsin (Promega, modified), the Coomassie-stained protein bands were excised and processed according to standard in-gel digestion practice.4 Desalting before mass spectrometry was carried out as described for the in-solution digests (below). Electroblotting. Gel-separated proteins were electroblotted to PVDF membrane (Novex, 0.2-µm pore size) using a semidry apparatus (Phase GmbH) operated at 0.8 mA/cm2 for 2 h at room temperature.16 The blot buffers are prepared and the blot stack is assembled according to the manufacturers recommendations (Phase GmbH). Briefly, cathode buffer is 25 mM Tris base containing 40 mM 6-aminocaproic acid and 20% methanol, while anode buffer I is 30 mM Tris base containing 20% methanol and anode buffer II is 300 mM Tris base containing 20% methanol. Following electroblotting, the PVDF-immobilized proteins were stained with Coomassie blue R-250 (0.1%, w/v) in 45% methanol and 2% acetic acid for 5-10 min. Destaining was carried out using 90% methanol containing 2% acetic acid for 0.5-1 h. Both staining and destaining were performed at room temperature after which the blots were dried and protein bands cut and stored individually in Eppendorf tubes at -20 °C until they were processed further (protein extraction, below). Protein Extraction. The PVDF-bound and Coomassie-stained protein band was divided into three pieces and placed in a 500µL plastic tube (Eppendorf). Extraction solvent (200 µL) containing 70% acetonitrile and 1% trifluoroacetic acid (TFA) was added to the tube followed by vortexing for 5 min. The sample was then sonicated for 30 min (Sonorex Super RK 514 BH, Bandelin, room temperature) and incubated at 42 °C overnight (shaking). Sonication for another 30 min followed, after which the solution was recovered and the PVDF pieces were rinsed with 50 µL of fresh (15) Laemmli, U. K. Nature 1970, 227, 680-685. (16) Kyhse-Andersen, J. J. Biochem. Biophys. Methods 1984, 10, 203-209.

Analytical Chemistry, Vol. 73, No. 22, November 15, 2001

5371

Figure 2. Schematic representations (drawn to scale) of the amino acid sequences of the proteins analyzed: (A) bovine carbonic anhydrase;18 (B) bovine actin;14 (C) Hrp 36 from C. tentans.13 The top position for each protein represents the possible tryptic fragments. The middle position represents portions of the sequence that were covered by mass spectrometry after protein extraction and in-solution digestion (shown by lines in bold). The coverage achieved after in-gel digestion is shown in the bottom position (dashed lines). T denotes tryptic fragment with number shown by index.

extraction solvent. The extraction solutions were pooled to yield a final volume of 250 µL that was dried under a stream of nitrogen. Amino Acid Analysis. To check the recovery of protein in the extraction procedure, PVDF membrane samples from before and after the acetonitrile/TFA extraction or aliquots of the protein extracts were placed in glass tubes for acid hydrolysis and amino acid analysis.17 The results were used to quantify the amount of protein immobilized onto the membrane, the amount that was left after extraction, or the protein content in the extract. From these data, the recovery in the extraction procedure was calculated. In-Solution Digestion. The dry protein extract was dissolved in 50 µL of 0.2 M ammonium bicarbonate (pH 7.8) and digested using 0.3 µg of modified trypsin (Promega) for 15 h at 37 °C. The reaction was quenched by freezing the sample. Desalting. Before mass spectrometric analysis, the samples were desalted on µ-C18 ZipTips (Millipore) and eluted in 75% acetonitrile containing 0.1% TFA for MALDI analysis and in 60% acetonitrile containing 1% acetic acid for ESI analysis. Mass Spectrometry. MALDI. Peptides, recovered in 0.8 µL of eluent from the desalting procedure (above), were mixed with an equal volume of R-cyano-4-hydroxycinnamic acid (saturated solution in 75% acetonitrile containing 0.1% TFA). The sample/ matrix mixture was applied to a stainless steel 100-well target plate (Applied Biosystems) and dried under a stream of air. A Voyager DE-Pro MALDI mass spectrometer (Applied Biosystems) operated in reflectron mode was used for the analysis. (17) Bergman, T.; Carlquist, M.; Jo ¨rnvall, H. Advanced Methods in Protein Microsequence Analysis; Springer-Verlag: Berlin, 1986; Chapter 1.4.

5372 Analytical Chemistry, Vol. 73, No. 22, November 15, 2001

ESI-MS/MS. Nano-ESI mass spectra were recorded using a quadrupole time-of-flight tandem mass spectrometer, Q-TOF (Micromass). The instrument was equipped with an orthogonal sampling ESI interface (Z-spray, Micromass). Metal-coated nanoESI needles (Protana) were used and manually opened on the stage of a light microscope to give a spraying orifice of ∼5 µm. This resulted in a flow of approximately 20-50 nL/min when a capillary voltage of 0.8-1.2 kV was applied. A nitrogen countercurrent drying gas facilitated desolvation. The cone voltage was optimized in the range 40-55 V. For the acquisition of collisioninduced dissociation (CID) spectra, the collision energy was optimized in the range 30-80 eV. Argon was used as the collision gas. RESULTS Three sets of proteins, Hrp 36, carbonic anhydrase, and two actins (one novel from C. tentans and the other from B. taurus (calf thymus)), were analyzed by MALDI and ESI mass spectrometry after gel separation, electroblotting, extraction, and insolution trypsin digestion. The same proteins and amounts were also in-gel digested, and the sequence coverage with each method was compared (Figure 2 and Table 1). Using the in-solution digestion method, protein modifications were identified by tandem mass spectrometry and the amino acid sequence of a novel actin isolated from C. tentans could be rapidly assessed in relation to the known actin sequences of Drosophila melanogaster, Mus musculus, and B. taurus.

Table 1. Percentage of Amino Acid Sequence Covered for Gel-Separated Proteinsa Using the Combined Data from MALDI and ESI Mass Spectrometry of Trypsin Digestsb sequence covered (%)

protein

in-gel digestion

in-solution digestion

improved coverage (%)

carbonic anhydrase Hrp 36 actin (bovine)

66.0 32.1 48.9

91.5 70.6 75.1

25.5 38.5 26.2

average

49.0

79.1

30.1

a A total of 2 pmol applied to gel, two to four experiments for each protein. b Analysis after standard in-gel digestion is compared to analysis after in-solution digestion of the protein recovered by electroblotting and extraction.

Proteins with a molecular mass up to ∼50 kDa were efficiently extracted from the PVDF membrane and could be recovered in solution. Generally, the protein extraction yield was 60-70% as determined by amino acid analysis but ranged from 43 up to 85%. At sizes above 50 kDa, as tested with bovine serum albumin (67 kDa) and phosphorylase B (97 kDa), the extraction is less efficient (data not shown), possibly due to stronger binding to the hydrophobic PVDF membrane. In this report, we focus on protein sizes up to 50 kDa. Protein amounts down to 1 pmol loaded onto the gel can be handled with good results in terms of detection and signal-to-noise ratio in the mass spectrometric analysis. The MALDI and ESI data for extracted and in-solution-digested carbonic anhydrase (30 kDa) allowed clear interpretation and extensive sequence coverage (in excess of 85%) for both 8 and 1 pmol of protein applied to gel electrophoresis (Figure 3). In ESI, the spectra revealed clear evidence for the presence of the 41-residue tryptic peptide T19 (quadruply charged at m/z 1149.0; cf. Figure 2) and the 25-residue T22 peptide (triply charged at m/z 951.5; cf. Figure 2). These two peptides, which are the largest tryptic peptides generated from carbonic anhydrase, were both absent in the spectra from the corresponding sample processed by in-gel digestion. CID of these peptides revealed their primary structures (for T19, see Figure 4). The carbonic anhydrase N-terminal tryptic peptide (m/z 1013.46; cf. Figure 3) was shown to be acetylated at the N-terminus by CID experiments. Data from Hrp 36 (40 kDa) and actin (42 kDa) revealed similar results, with large tryptic peptides absent in the mass maps when in-gel digestion was used. In contrast, when digestion was carried out in-solution after protein extraction from the PVDF membrane, the large tryptic peptides were in general present in the MALDI mass spectrum and could be analyzed by CID experiments (Figure 2). A comparison between in-gel digestion and in-solution digestion of extracted bovine actin shows clearly that high-molecular-weight tryptic peptides are more abundant in the MALDI spectra from the solution method (Figure 5). The Hrp 36 protein is not ideally suited for tryptic digestion since the primary structure is cleaved into peptides of widely different sizes. A 67-residue peptide (T33; cf. Figure 2) with a molecular weight of 7091.8 is generated in addition to a number of small peptides (less than five residues) because of the scattered density of basic residues in the Hrp 36

sequence. The 67-residue peptide corresponds to almost 23% of the entire protein sequence and could not be detected after ingel digestion. However, after extraction from PVDF and digestion in-solution, starting from 2 pmol of Hrp 36 applied to the gel, it was possible to detect this peptide in the MALDI mass spectrum (Figure 2). To achieve this, linear time-of-flight detection optimized for larger polypeptides than normally appear in the analysis of tryptic digests was used. The data were still good enough to clearly detect oxidation of Met at position 45 in this peptide. Amino acid sequence coverage of all three proteins tested is on the average 49% after in-gel digestion while the current protocol with protein extraction followed by in-solution digestion generates an average sequence coverage of 79%, an increase of 30% (Table 1). The extraction and in-solution digestion protocol was applied to noncharacterized actin (no protein or DNA sequence available) from the dipteran C. tentans where ∼10 pmol of protein was applied to the gel. A large part of the novel sequence (>75%) could be rapidly assessed using MALDI mass mapping of tryptic fragments complemented by tandem mass spectrometry. A few amino acid substitutions versus other actins were identified from the CID data, as well as a posttranslational methylation of a His residue at position 5 in T8. In addition, an artifactual acrylamide modification of Cys at the penultimate position of T1 was detected (Figure 6). DISCUSSION Conventional mass spectrometric analysis of proteins frequently involves in-gel digestion steps.4,5 However, large proteolytic peptides are often difficult to recover and analyze. For example, the two largest tryptic peptides of carbonic anhydrase, T19 and T22, were not possible to detect in the spectra from the samples processed by in-gel digestion. This indicates that large proteolytic peptides are trapped in the gel and are not eluted, resulting in a less good sequence coverage compared to the suggested method. In contrast, both peptides T19 and T22 were recovered using in-solution digestion, indicating the efficiency of the current approach when a reasonable amount of sample is available (1-10 pmol applied to gel electrophoresis) and a good sequence coverage is necessary. This is particularly true in the search for posttranslational modifications and in the differentiation between protein isoforms. For carbonic anhydrase, acetylation of the N-terminus could be shown via tandem mass spectrometry of the corresponding tryptic peptide which was no problem to detect after in-solution digestion. Similar results were obtained with the other two proteins tested. Four additional large peptides were recovered by applying the in-solution method on bovine actin compared to use of in-gel digestion (Figure 2). For Hrp 36, a 67-residue peptide corresponding to almost 23% of the primary structure was recovered and detected after protein extraction followed by digestion in solution. This peptide, together with several of the other tryptic peptides, was absent using in-gel digestion, which thus gives a less good sequence coverage. The average protein recovery in the extraction step was estimated to be within 60-70% based on the amino acid analysis data. Either the amount of membrane-bound protein before and after extraction was determined or the amount of protein in the extract was analyzed and compared to the amount remaining on Analytical Chemistry, Vol. 73, No. 22, November 15, 2001

5373

Figure 3. MALDI mass spectra of carbonic anhydrase after electroblotting, extraction from the membrane and trypsin in-solution digestion, starting from (A) 8 and (B) 1 pmol of protein applied to the electrophoresis gel. 5374

Analytical Chemistry, Vol. 73, No. 22, November 15, 2001

Figure 4. Collision-induced dissociation spectrum of the [M + 4H]4+ ion of tryptic fragment T19 from in-solution digestion of carbonic anhydrase (cf. Figure 2). (The Roepstorff and Fohlman nomenclature19 was originally designed to describe singly charged product ions. It is therefore implicit in our spectra that for doubly charged product ions an additional proton has been added.)

the membrane after extraction. However, we do not specifically include the electroblotting yield since it is difficult to separate from the protein losses that occur during application of sample to the gel and in the electrophoresis process. Other conditions for protein extraction were initially tested: higher acetonitrile percentage, 2-propanol instead of acetonitrile, and higher TFA concentration. Each of the variant conditions resulted in a lower protein recovery compared to that achieved using the current protocol (i.e., 70% acetonitrile and 1% TFA). It should be pointed out that the extraction and in-solution digestion method described here cannot compete with in-gel digestion in terms of speed (high throughput) and sensitivity in protein identifications. Nevertheless, using a reasonable amount of sample (stainable by Coomassie blue) and having specific questions regarding the primary structure and its modifications, this method offers better sequence coverage and structural characterization than the corresponding in-gel digestion approach. In two-dimensional gel electrophoresis, series of spots with similar molecular weights but with significantly different pI values are often seen. These spots usually reveal the same protein identity after in-gel digestion, peptide mass mapping, and database searches. Presumably, the different pI values are due to minor truncations or to modifications such as phosphorylation, sulfation, and deamidation. These modifications are generally difficult to

detect when the sequence coverage is less than 50% such as after standard in-gel digestion. The chance of detection naturally increases when a larger portion of the sequence is covered as in the present method with extraction of intact protein before digestion. The fact that the digestion is carried out in-solution results in improved recovery of large proteolytic peptides and, hence, in an overview of a greater part of the primary structure with better possibilities to detect the critical peptides that are modified. However, it should be noted that signals from peptides of less than five residues are hard to distinguish from the high background in that spectrum region. Some larger peptides can also be difficult to detect in mixtures since ESI and MALDI are competitive ionization processes. Methylation of a single His residue in actin from the dipteran C. tentans was demonstrated by CID of the corresponding tryptic peptide. This protein modification is not commonly found but has been described.22 We found the methylation in peptide T8 of the (18) Sciaky, M.; Limozin, N.; Filippi-Foveau, D.; Gulian, J. M.; Laurent-Tabusse, G. Biochimie 1976, 58, 1071-1082. (19) Roepstorff, P.; Fohlman, J. Biomed. Mass Spectrom. 1984, 11, 601. (20) Fyrberg, E. A.; Bond, B. J.; Hershey, N. D.; Mixter, K. S.; Davidson, N. Cell 1981, 24, 107-116. (21) Peter, B.; Man, Y. M.; Begg, C. E.; Gall, I.; Leader, D. P. J. Mol. Biol. 1988, 203, 665-675. (22) Johnson, P.; Harris, C. I.; Perry, S. V. Biochem. J. 1967, 105, 361-370.

Analytical Chemistry, Vol. 73, No. 22, November 15, 2001

5375

Figure 5. Comparison of MALDI mass spectra for actin digests (2 pmol applied to gel electrophoresis) in the high-molecular-weight region (2700-3400 m/z range). The trypsin digests were prepared in-solution after protein extraction (A) and in the gel (B). 5376 Analytical Chemistry, Vol. 73, No. 22, November 15, 2001

Figure 6. Alignment of the partially determined novel actin sequence from Chironomus tentans (MALDI and tandem mass spectrometry of tryptic peptides from in-solution digestion) with the known amino acid sequences of Drosophila melanogaster,20 Mus musculus,21 and Bos taurus14 actins. Conserved amino acid residues are indicated by asterisks (*). Residue substitutions to similar amino acids are indicated by colons (:). Dashed lines indicate portions of the C. tentans actin sequence that were not analyzed, and X denotes an unidentified residue. The methylated His and the acrylamide derivatized Cys in C. tentans actin are indicated by filled squares.

novel actin from C. tentans (Figure 6). For this, clearly 1 out of 36 tryptic peptides must be found and analyzed by CID, emphasizing the importance of as good sequence coverage as possible by use of in-solution digestion. Yet another benefit of using digestion in-solution is that a wider selection of proteinases can be made compared to in-gel approaches, increasing flexibility regarding proteolytic specificity. In conclusion, the proposed method for recovery of intact proteins from gel separations for digestion in-solution (Figure 1) offers an alternative to the standard in-gel digestion procedure and gives improved sequence coverage and better possibilities to detect functionally important protein modifications.

ACKNOWLEDGMENT This work was supported by grants from the Swedish Research Council (projects 03X-3532, 03X-10832, B5101-879/2001, and K5104-20005891), the Swedish Cancer Society (project 4159), the Foundation for Strategic Research (Cell Factory), the European Commission (BIO4-CT97-2123), and the Swedish Society of Medicine.

Received for review April 30, 2001. Accepted August 2, 2001. AC010486H

Analytical Chemistry, Vol. 73, No. 22, November 15, 2001

5377