Recombinant Isotope Labeled and Selenium Quantified Proteins for

These standards are recombinant proteins containing an isotope label and selenomethionine. For recombinant protein expression, assembly of expression ...
0 downloads 4 Views 1MB Size
Download PDF
Anal. Chem. 2010, 82, 2334–2340

Recombinant Isotope Labeled and Selenium Quantified Proteins for Absolute Protein Quantification Nico Zinn, Dominic Winter,† and Wolf D. Lehmann* Molecular Structure Analysis, German Cancer Research Center, Im Neuenheimer Feld 280, Heidelberg, Germany A novel, widely applicable method for the production of absolutely quantified proteins is described, which can be used as internal standards for quantitative proteomic studies based on mass spectrometry. These standards are recombinant proteins containing an isotope label and selenomethionine. For recombinant protein expression, assembly of expression vectors fitted to cell-free protein synthesis was conducted using the gateway technology which offers fast access to a variety of genes via open reading frame libraries and an easy shuttling of genes between vectors. The proteins are generated by cell-free expression in a medium in which methionine is exchanged against selenomethionine and at least one amino acid is exchanged by a highly stable isotope labeled analogue. After protein synthesis and purification, selenium is used for absolute quantification by element mass spectrometry, while the heavy amino acids in the protein serve as reference in subsequent analyses by LC-ESI-MS or MALDI-MS. Accordingly, these standards are denominated RISQ (for recombinant isotope labeled and selenium quantified) proteins. In this study, a protein was generated containing Lys+6 ([13C6]-lysine) and Arg+10 ([13C6,15N4]-arginine) so that each standard tryptic peptide contains a labeled amino acid. Apolipoprotein A1 was synthesized as RISQ protein, and its use as internal standard led to quantification of a reference material within the specified value. Owing to their cellfree expression, RISQ proteins do not contain posttranslational modifications. Thus, correct quantitative data by ESI- or MALDI-MS are restricted to quantifications based on peptides derived from unmodified regions of the analyte protein. Therefore, besides serving as internal standards, RISQ proteins stand out as new tools for quantitative analysis of covalent protein modifications. Method development for quantification is an important focus in current proteomic research, since quantitative proteomic data can provide new insights, e.g., into the balancing of metabolic * Corresponding author. Address: Molecular Structure Analysis (W160), German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. E-mail: [email protected]. Phone: +49-6221-42 45 63. Fax: +496221-42 56 61. † Current address: Department of Pathology, Harvard Medical School and Children’s Hospital, Boston, MA 02115.

2334

Analytical Chemistry, Vol. 82, No. 6, March 15, 2010

pathways or the architecture and cooperativity of signaling processes. For relative comparison of protein concentrations between two or more samples, several mass spectrometric methods have been developed. Quantitative data with good accuracy can be generated in particular by methods employing stable isotope labeling. These methods carry the acronyms iCAT,1 SILAC,2 TMT,3 iTRAQ,4 or ICPL.5 Whereas SILAC uses metabolic and the methods iCAT, TMT, iTRAQ, and ICPL use chemical labeling and differential digestion in 16O/18O, water employs in vitro enzymatic labeling.6 These methods have been reviewed repeatedly in the last years.7-11 Even more relevant than measuring relative quantitative changes of proteins is the determination of their concentrations, denominated in this context as absolute quantification. Such absolute quantifications are important, e.g., for biomarker detection and validation. Moreover, the availability of protein concentration data allows the development and testing of models describing in a simplified way essential features of complex cellular processes. While the actual analytical steps by mass spectrometry are identical for relative and absolute quantifications, the latter is more challenging, since it requires the availability of absolutely quantified peptide or protein standards. The strategies developed so far for absolute proteomic quantifications by mass spectrometry are summarized in Figure 1. Absolute peptide quantification is connected with stable isotope labeled peptides denominated as AQUA,12 PolySIS,13 QconCAT,14 (1) Gygi, S. P.; Rist, B.; Gerber, S. A.; Turecek, F.; Gelb, M. H.; Aebersold, R. Nat. Biotechnol. 1999, 17, 994–999. (2) Ong, S.-E.; Blagoev, B.; Kratchmarova, I.; Kristensen, D. B.; Steen, H.; Pandey, A.; Mann, M. Mol. Cell. Proteomics 2002, 1, 376–386. (3) Thompson, A.; Schafer, J.; Kuhn, K.; Kienle, S.; Schwarz, J.; Schmidt, G.; Neumann, T.; Hamon, C. Anal. Chem. 2003, 75, 1895–1904. (4) Ross, P. L.; Huang, Y. N.; Marchese, J. N.; Williamson, B.; Parker, K.; Hattan, S.; Khainovski, N.; Pillai, S.; Dey, S.; Daniels, S.; Purkayastha, S.; Juhasz, P.; Martin, S.; Bartlet-Jones, M.; He, F.; Jacobson, A.; Pappin, D. J. Mol. Cell. Proteomics 2004, 3, 1154–1169. (5) Alexander, S.; Josef, K.; Friedrich, L. Proteomics 2005, 5, 4–15. (6) Schnolzer, M.; Jedrzejewski, P.; Lehmann, W. D. Electrophoresis 1996, 17, 945–953. (7) Heck, A. J. R.; Krijgsveld, J. Expert Rev. Proteomics 2004, 1, 317–326. (8) Mayya, V.; Han, D. K. Expert Rev. Proteomics 2006, 3, 597–610. (9) Putz, S.; Reinders, J.; Reinders, Y.; Sickmann, A. Expert Rev. Proteomics 2005, 2, 381–392. (10) Bantscheff, M.; Schirle, M.; Sweetman, G.; Rick, J.; Kuster, B. Anal. Bioanal. Chem. 2007, 389, 1017–1031. (11) Miyagi, M.; Rao, K. C. S. Mass Spectrom. Rev. 2007, 26, 121–136. (12) Gerber, S. A.; Rush, J.; Stemman, O.; Kirschner, M. W.; Gygi, S. P. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 6940–6945. 10.1021/ac9025412  2010 American Chemical Society Published on Web 02/17/2010

Figure 1. Summary of absolute quantification methods using mass spectrometry-based proteomics: (a) for peptides and (b) for proteins.

or PASTA15 standards. AQUA peptides are quantified by amino acid analysis (AAA), whereas PASTA peptides are quantified by LC-element mass spectrometry. These single peptide standards can be added directly before or after the protein digestion step. In both the PolySIS and QconCAT concept, several labeled standard peptides are synthesized in a concatenated form as a single protein which is purified and then quantified by AAA. Compared to the use of single AQUA peptides, the benefit of the PolySIS/QconCAT concept is the production and quantification of several standard peptides in a single step. Consequently, multiple proteins can be quantified by a single PolySIS/QconCAT standard. However, such standards of course cannot mimic the analytical behavior of the intact individual proteins, for which the concatenated standard is designed. The ideal scenario for absolute protein quantification is the addition of an absolutely quantified and isotopically labeled, but otherwise identical protein analogue, at the start of the analytical process. For relative quantification, metabolic incorporation of stable isotope labeled amino acids offers a close-to-perfect solution, e.g., for cell culture experiments. It has been attempted to transfer this concept to absolute quantification.16 However, the concept of metabolic label incorporation loses its general value in the context of absolute quantification, since individual proteins have to be isolated before quantification. This can be achieved by the established techniques of cell-free biosynthesis of a protein with recombinant His-tag in combination with Ni-column purification or by immunoprecipitation.16,17 As an alternative, standard-free relative quantification within a labeled proteome can be performed, e.g., using the emPAI18 score in combination with a quantification of the total protein concentration. The latter approach is generally applicable, but it delivers much less accurate quantitative data compared to isotope dilution. As a variation of cell-free synthesis designed for protein quantification, the FLEXIQuant method has recently been introduced, which employs a protein construct (13) Anderson, L.; Hunter, C. L. Mol. Cell. Proteomics 2006, 5, 573–588. (14) Pratt, J. M.; Simpson, D. M.; Doherty, M. K.; Rivers, J.; Gaskell, S. J.; Beynon, R. J. Nat. Protoc. 2006, 1, 1029–1043. (15) Zinn, N.; Hahn, B.; Pipkorn, R. d.; Schwarzer, D.; Lehmann, W. D. J. Proteome Res. 2009, 8, 4870–4875. (16) Hanke, S.; Besir, H.; Oesterhelt, D.; Mann, M. J. Proteome Res. 2008, 7, 1118–1130. (17) Brun, V.; Dupuis, A.; Adrait, A.; Marcellin, M.; Thomas, D.; Court, M.; Vandenesch, F.; Garin, J. Mol. Cell. Proteomics 2007, 6, 2139–2149. (18) Ishihama, Y.; Oda, Y.; Tabata, T.; Sato, T.; Nagasu, T.; Rappsilber, J.; Mann, M. Mol. Cell. Proteomics 2005, 4, 1265–1272.

elongated by an enzyme-cleavable peptide.19 As surrogate for the complete protein, this peptide can be absolutely quantified relative to a peptide standard following proteolysis. In the following study, an innovative method for absolute protein quantification is introduced, which uses the incorporation of selenomethionine into a recombinantly expressed protein as quantification tag. EXPERIMENTAL PROCEDURES Production of Expression Vector. The expression vector pEXP1-DEST, LR-clonase mix, and TOP10 chemically competent E. coli were from Invitrogen (Karlsruhe, Germany). The gateway entry clone pDONR221 containing the Apolipoprotein A1 gene sequence was derived from an in-house clone repository. The transfer of the ApoA1 gene was carried out using the LR-clonase mix20 as suggested by the manufacturer, and the construct was transformed into TOP10 chemically competent E. coli which were transferred to agar-plates with LB broth (Roth, Karlsruhe, Germany) containing 100 µg mL-1 ampicillin (Sigma Aldrich, Taufkirchen, Germany) and incubated overnight. Single colonies were picked and incubated overnight in 1 mL of LB broth. The expression vector was purified using the QIAprep mini preparation kit (Qiagen, Hilden, Germany). The vector sequence was verified by in-house sequencing, and E. coli strains containing the verified sequence were further propagated in 100 mL of LB medium. The vector was purified with a Jetstar plasmid preparation kit (Genomed, Lo¨hne, Germany) resulting in 600 µg of purified expression vector per preparation. Cell-Free Protein Expression. The protein was expressed using the Expressway cell-free E. coli expression system (Invitrogen, Karlsruhe, Germany) in 200 µL portions within 6 h using the manufacturer’s protocol but with a special amino acid mixture which contained all proteinogenic L-amino acids except for Lys, Arg, and Met. These were exchanged against L-[13C6,15N4]arginine, L-[13C6]-lysine, and L-selenomethionine, respectively. All amino acids were obtained as hydrochloride from Sigma Aldrich (Taufkirchen, Germany), and a mixture containing each amino acid in a concentration of 50 mM in 50 mM 4-(2Hydroxyethyl)piperazine-1-ethanesulfonic acid (Hepes) (Sigma, Taufkirchen, Germany) buffer pH 11 was prepared. The selenomethionine solution was prepared differently at 75 mM in 50 mM Hepes (pH 7.5) with 4 mM dithiothreitol (DTT) (19) Singh, S.; Springer, M.; Steen, J.; Kirschner, M. W.; Steen, H. J. Proteome Res. 2009, 8, 2201–2210. (20) Landy, A. Annu. Rev. Biochem. 1989, 58, 913–949.

Analytical Chemistry, Vol. 82, No. 6, March 15, 2010

2335

(Sigma, Taufkirchen, Germany). After expression, the protein was stored at -20 °C until further use. RISQ Protein Purification. Purification was carried out using the His-tag technology. Therefore, 200 µL of cell-free expression system was incubated with 400 µL of cellytic inclusion body solubilization reagent (Sigma, Taufkirchen, Germany) for 30 min. Then, 400 µL of 8 M urea in 50 mM sodium phosphate buffer, 300 mM sodium chloride, 10 mM imidazole (pH 7.4, Nunc, Langenselbold, Germany) were added, and the pH of the solution was adjusted with either hydrochloric acid or sodium hydroxide to pH 7.4. Samples were 0.22 µm filtered using centrifugal units (Millipore, Schwalbach, Germany) and applied to Nickel metalchelate spin columns (Nunc, Langenselbold, Germany). All centrifugation steps were performed at 4 °C in a refrigerated microcentrifuge (Hermle, Wehingen, Germany). Washing steps were carried out using 50 mM sodium phosphate buffer, 300 mM sodium chloride, and 50 mM imidazole (pH 7.4), and the protein was eluted with 50 mM sodium phosphate buffer, 300 mM sodium chloride, and 400 mM imidazole (pH 7.4). To remove imidazole and elution buffer, the eluted protein fraction was dialyzed against 10 mM ammonium hydrogen carbonate using 6-8 kDa MWCmembrane tubes (GeBAflex, Hahalon, Israel) for 24 h. The purification was checked by SDS-gel electrophoresis using precast 4-12% Tris-glycine gels and 3-[N-morpholino] propanesulfonic acid running buffer (Invitrogen, Karlsruhe, Germany). Protein Samples. The certified reference material CRM 393 with a specified content in ApoA1 (accession no. P02647) produced by the Institute of Reference Materials and Measurements (Geel, Belgium) was from Sigma (Taufkirchen, Germany). The reference material was reconstituted as prescribed to obtain a solution with 37.7 ± 1.8 µmol L-1 of ApoA1 and directly applied for analysis. The surplus solution was aliquoted and stored at -20 °C until further use. Apolipoprotein A1 from human plasma (1.1 mg mL-1) was from Sigma. Enzymatic Digestion. For protein identification and purity control, the purified protein was subjected to in-solution and ingel21 digestion without reduction and alkylation overnight using sequencing grade trypsin and/or AspN (Roche, Mannheim, Germany). The in-solution digestion was carried out using a final concentration of 1 M urea (Sigma, Taufkirchen, Germany) in the digestion buffer as denaturant. Liquid Chromatography Mass Spectrometry. LC-MS/MS analyses of the protein digest were carried out using a UPLCsystem (Waters, Milford, USA) equipped with a 75 µm × 150 mm C18 column with 1.7 µm particle size (Waters, Milford, USA) connected to either a QTOF (Waters Micromass, Manchester, UK) or a LTQ-Orbitrap XL system (Thermo, Bremen, Germany). Mascot (version 2.2.0.4) and Mascot Distiller Quantitation Toolbox (version 2.1.3.0) (Matrix Science, London, UK) were used for protein identification and automated quantification. Inductively Coupled Plasma Mass Spectrometry. The HPLC separations were performed on a capLC system (Waters, Milford, MA) equipped with a 300 µm × 150 mm capillary column (Waters, Milford, MA) packed with C4 particles of 5 µm and a pore size of 300 Å coupled online to a sector field ICPMS type Element2 (Thermo, Bremen, Germany) through a low-flow mi(21) Rosenfeld, J.; Capdevielle, J.; Guillemot, J. C.; Ferrara, P. Anal. Biochem. 1992, 203, 173–179.

2336

Analytical Chemistry, Vol. 82, No. 6, March 15, 2010

croconcentric nebulizer CEI-100 (CETAC, Omaha, USA). Flow injections were performed with a microconcentric PFA-nebulizer (Elemental Scientific, Omaha, USA) and a Scott-type spray chamber (AHF Analysentechnik, Tu¨bingen, Germany) with selenium standard solution 1000 ± 2 mg L-1 (Sigma, Taufkirchen, Germany). The ICPMS was operated with the following conditions: sample gas, 1.0 L min-1; auxiliary gas, 0.8 L min-1; plasma power, 1300 W. 82Se was monitored with a mass spectrometric resolution of 300 and a dwell time of 200 msec. Data evaluation was performed using Origin 8G (Northhampton, USA). RESULTS AND DISCUSSION Concept. We developed a method for the generation of absolutely quantified protein standards based on cell-free synthesis and their quantification by element mass spectrometry (ICPMS). Protein quantification via detection of sulfur present in Cys and Met residues is possible, as demonstrated also for reference ApoA1.22 However, sulfate present in solvents and in particular in sulfur-containing additives to biochemical buffers (e.g., DTT) limits this approach. We selected the introduction of selenomethionine as a quantification tag for ICPMS, since (i) it can quantitatively replace methionine using cell-free protein synthesis,23 since (ii) selenium background is generally low, and because (iii) element mass spectrometry shows a high sensitivity for selenium. The application of cell-free protein synthesis24,25 in combination with substitution of selected amino acids by stable isotope labeled analogs26,27 has been demonstrated. The exchange of L-methionine by L-selenomethionine by in vitro translation23 was first explored in connection with NMR and X-ray diffraction analysis.28 The introduction of 34S-Met, Se-Met, and Te-Met was also demonstrated, and it was observed that 34S-Met and SeMet showed a similar stability whereas Te-Met was unstable.29 On this basis, we selected to employ combined incorporation of stable isotope labeled amino acids and selenomethionine for preparation of standard proteins as schematically shown in Figure 2 and to name these proteins recombinant isotopically labeled and selenium quantified (RISQ) proteins. The presence of selenomethionine in RISQ proteins serves as a quantification tag for ICPMS, whereas stable isotope labeling allows quantification of the target protein by molecular mass spectrometry relative to the absolutely quantified standard. For methionine-containing peptides of the analyte protein, the standard protein delivers the selenomethionine-incorporated analogues, which are not ideally suited as standards due to their different LC-retention times and possible ionization efficiency differences. (22) Zinn, N.; Kru ¨ ger, R.; Leonhard, P.; Bettmer, J. Anal. Bioanal. Chem. 2008, 391, 537–543. (23) Kigawa, T.; Yamaguchi-Nunokawa, E.; Kodama, K.; Matsuda, T.; Yabuki, T.; Matsuda, N.; Ishitani, R.; Nureki, O.; Yokoyama, S. J. Struct. Funct. Genomics 2002, 2, 29–35. (24) Katzen, F.; Chang, G.; Kudlicki, W. Trends Biotechnol. 2005, 23, 150–156. (25) Murthy, T. V. S.; Wu, W.; Qiu, Q. Q.; Shi, Z.; LaBaer, J.; Brizuela, L. Protein Expression Purif. 2004, 36, 217–225. (26) Kigawa, T.; Yabuki, T.; Yoshida, Y.; Tsutsui, M.; Ito, Y.; Shibata, T.; Yokoyama, S. FEBS Lett. 1999, 442, 15–19. (27) Matsuda, T.; Koshiba, S.; Tochio, N.; Seki, E.; Iwasaki, N.; Yabuki, T.; Inoue, M.; Yokoyama, S.; Kigawa, T. J. Biomol. NMR 2007, 37, 225–229. (28) Kiyoshi, O.; Nicholas, E. D.; Gottfried, O. IUBMB Life 2005, 57, 615–622. (29) Ogra, Y.; Kitaguchi, T.; Suzuki, N.; Suzuki, K. Anal. Bioanal. Chem. 2008, 390, 45–51.

Figure 2. Concept of RISQ proteins. Two different kinds of labels are included by in vitro translation, selenomethionine substitutes for methionine as an absolute quantification tag for ICPMS and stable isotope labeled amino acids serving as standards for molecular mass spectrometrybased proteomics.

Figure 3. 1D-gel separation of His-tagged Apolipoprotein A1 generated by cell-free synthesis. Left: Coomassie stained 1D gel: lane 1, marker proteins; lane 2, total protein from the synthesis system; lane 3, Ni-column flow-through; lane 4, elution fraction from Ni-column showing ApoA1 (35.7 kDa). Right: Sequence of ApoA1 with analytical sequence coverage by LC-MS/MS (underlined).

However, the majority of peptides are not affected by incorporation of selenomethionine due to the low natural abundance of methionine. Preparation of the Cell-Free Expression Vector. We selected Apolipoprotein A1 as model protein for cell-free synthesis. For this purpose, the ApoA1 gene was integrated in the expression vector pEXP1-DEST containing a T7 promotor30 and a ribosome binding site. The gene was transferred using the gateway expression system31 offering access to about 20 000 commercially available genes. After transfer and selection, the desired structure of the pEXP1-DEST vector was obtained as verified by sequencing. Cell-Free Protein Expression. Cell-free protein expression was performed over 6 h in a medium containing all standard proteinogenic amino acids except methionine which was exchanged against selenomethionine. In addition, Lys and Arg were exchanged against Lys+6 [13C6-Lys] or Arg+10 (30) Studier, F. W.; Rosenberg, A. H.; Dunn, J. J.; Dubendorff, J. W. Methods Enzymol. 1990, 185, 60–89. (31) Esposito, D.; Garvey, L. A.; Chakiath, C. S. In High throughput protein expression and purification; Doyle, S. A., Ed.; Springer: Heidelberg, 2009; Vol. 498, p 31-54.

[13C6,15N4], respectively. The newly synthesized protein was extracted via a standard His-tag purification, which resulted in a single Coomassie-stainable protein band as demonstrated in Figure 3. For further purification of the protein, the eluate fraction from the Ni-column was subjected to dialysis against 10 mM ammonium hydrogen carbonate and in-solution digestion by trypsin. LC-MS/ MS analysis and database searching resulted in a sequence coverage of 87% for ApoA1. The major missing part was a short N-terminal region containing the His-tag (see Figure 3). This failure can be ascribed to the fact that this peptide exhibits an extremely unfavorable elution behavior in reversed phase LC, as confirmed by a synthetic peptide containing a His-tag (not shown). Inspection of the MS and MS/MS spectra confirmed incorporation of at least one residue of Lys+6 or Arg+10 into the tryptic peptides as well as incorporation of selenomethionine instead of methionine. The presence of selenium can be easily recognized by the complex isotope pattern, as demonstrated in Figure 4 by comparison of the experimentally detected and the simulated isotopic pattern of the molecular ion of WQEE-SeM-ELYR. Analytical Chemistry, Vol. 82, No. 6, March 15, 2010

2337

Figure 4. Mass spectrometric analysis of the tryptic peptide WQEE-SeM-ELYR: upper panel, simulated isotope pattern of the [M + 2H]2+ ion; lower panel, observed isotope pattern of the [M + 2H]2+ ion (SeM ) selenomethionine).

Figure 5. Analysis of purified RISQ Apoliprotein A1 by LC-ICPMS and 82Se detection. The solitary selenium peak indicates the elution of RISQ ApoA1 as the sole selenium containing species.

For peptides observed at high abundance, the presence of up to 3% of unlabeled analogues was observed, which indicated the presence of a minor amount of unlabeled free Lys and Arg within the cell-free synthesis system. For peptide signals observed at medium to low abundance, including all Se-Met containing peptides, no such unlabeled satellite ions were observed. Quantification of RISQ ApoA1. Following confirmation of its identity, the RISQ ApoA1 standard was quantified via selenium detection using ICPMS. To demonstrate the presence of only a single selenium species within the purified fraction, a LC-ICP MS run was conducted (Figure 5). The analysis revealed the absence of residual free selenomethionine and the presence of only a single protein. Thus, the RISQ protein was quantified by flow injection using an inorganic selenium standard as reference. In this way, further data processing steps such as LC-gradient correction of 2338

Analytical Chemistry, Vol. 82, No. 6, March 15, 2010

the Se ionization efficiency32 and a separate LC recovery determination of the RISQ protein are avoided. Taking into account the presence of eight selenomethionine residues in the sequence of recombinant ApoA1, the seleniumbased quantification leads to a concentration of 97 ± 11 fmol µL-1 of RISQ ApoA1. To prove the correctness of the quantitation, the recombinant protein was mixed with a certified reference material for ApoA1 available from the Institute for Reference Materials and Measurements (IRMM, Geel, Belgium) with concentration of 37.7 ± 1.8 µmol L-1. Reference ApoA1 was diluted to a concentration of 94.3 fmol µL-1 and mixed with RISQ ApoA1 of 97 fmol µL-1, resulting in an expected molar ratio of 0.97: 1. Following tryptic digestion, quantification was (32) Allain, P.; Jaunault, L.; Mauras, Y.; Mermet, J. M.; Delaporte, T. Anal. Chem. 1991, 63, 1497–1498.

Figure 6. Ratio of reference ApoA1 to RISQ ApoA1 determined by ratio analysis of nonlabeled to labeled tryptic fragments (light/heavy). The peptide ratios are displayed in the order of their position in the protein chain. The overall normalized ratio of 0.93 ( 0.2 is identical to the expected value within the error of the analysis.

Figure 7. Ratio of native ApoA1 to RISQ ApoA1 determined by ratio analysis of nonlabeled to labeled tryptic fragments (light/heavy). The ratios are displayed in the order of the position of the peptides in the protein chain. For peptides from the central part of the protein, rough agreement between the expected and experimental ratio is observed. For calculation of the normalized mean ratio (dashed line), the two N-terminal and the three C-terminal peptides were excluded. Their strongly deviating ratios indicate modifications in the native protein. Analytical Chemistry, Vol. 82, No. 6, March 15, 2010

2339

Figure 8. Analysis of a mixture of native ApoA1 and RISQ ApoA1 in an expected ratio of 0.95 to 1 by LC-ESI-MS. Left panel: molecular ion group of nonlabeled and labeled AHVDALR; right panels: extracted ion traces for the molecular ion group of the nonlabeled and labeled peptide, respectively. The light/heavy ratio for this peptide shows good agreement with the expected ratio.

performed by LC-MS. The first two peptides from the Nterminus representing the ApoA1 signal peptide were virtually absent in reference ApoA1 due to a cleavage of the signal peptide in vivo.33 Consequently, these peptides were not considered for the calculation of the mean. Analysis of all other peptide ratios by LC-MS (light/heavy) resulted in an average ratio of reference ApoA1 to RISQ ApoA1 of 0.91 ± 0.18 (n ) 29 peptides, SD; four replicates), which is in agreement with the expected ratio. For facilitated recognition of the deviation between expected and experimental data, the experimental data were normalized to the expected ratio of 0.97 to achieve a graphical display with the expected ratio at 1.0. This procedure resulted in an experimental ratio of 0.93 ± 0.2 as displayed in Figure 6. Another commercial sample of ApoA1 from human plasma which is denominated native ApoA1 was analyzed in addition. The native ApoA1 was mixed with RISQ ApoA1 to an expected ratio of 0.95:1. However, using all peptides except for the two signal peptides at the N-terminus, an experimental mean value of 0.70 ± 0.2 (n ) 16 peptides, SD) was observed. By inspection of the individual peptide ratios as displayed in Figure 7, it became evident that most values were close to the expected ratio, as for example that observed for the peptide AHVDALR (see Figure 8). As observed for reference ApoA1, (33) Brewer, H. B.; Fairwell, T.; Kay, L.; Meng, M.; Ronan, R.; Law, S.; Light, J. A. Biochem. Biophys. Res. Commun. 1983, 113, 626–632. (34) Beg, Z. H.; Stonik, J. A.; Hoeg, J. M.; Demosky, S. J., Jr.; Fairwell, T.; Brewer, H. B., Jr. J. Biol. Chem. 1989, 264, 6913–6921. (35) Hoeg, J. M.; Meng, M. S.; Ronan, R.; Fairwell, T.; Brewer, H. B., Jr. J. Biol. Chem. 1986, 261, 3911–3914. (36) Lemieux, J.; Giannoulis, S.; Breckenridge, W. C.; Mezei, C. Neurochem. Res. 1995, 20, 269–278. (37) Moore, L. E.; Fung, E. T.; McGuire, M.; Rabkin, C. C.; Molinaro, A.; Wang, Z.; Zhang, F.; Wang, J.; Yip, C.; Meng, X.-Y.; Pfeiffer, R. M. Cancer Epidemiol. Biomarkers Prev. 2006, 15, 1641–1646.

2340

Analytical Chemistry, Vol. 82, No. 6, March 15, 2010

native ApoA1 also showed the absence of the two most N-terminal peptides, indicating cleavage of the signal peptide. In addition, in native ApoA1, the three most C-terminal peptides also showed extremely low light/heavy ratios. This is consistent with the known C-terminal modifications of ApoA1,34-37 which are absent in RISQ ApoA1. Therefore, the deviating ratios were excluded from the calculation of the average which then gave a mean ratio of 0.87 ± 0.2 (n ) 13 peptides, SD), a value which is still in agreement with the expected ratio. The slight deviation from the expected ratio may be caused by the presence of impurities in the commercial peptide sample, and/or the increase of the molecular weight of native ApoA1 by covalent modifications. In summary, RISQ proteins can be generated by commercially available kits for cloning and cell-free protein synthesis. They are quantified by element mass spectrometry based on selenomethionine incorporation and, thus, provide accurately quantified protein standards. By concept, RISQ proteins allow the close approximation between standard and target proteins by cell-free biosynthesis, since the quantification tag is inherent in the sequence in the form of selenomethionine. ACKNOWLEDGMENT The authors are indebted to S. Bohl and U. Klingmu¨ller for valuable support and to A. Leischwitz and B. Korn for providing the ApoA1 clone. This work was supported in part by the Cooperation Program in Cancer Research of the German Cancer Research Center (DKFZ) and Israeli’s Ministry of Science and Technology (MOST).

Received for review November 6, 2009. Accepted February 3, 2010. AC9025412