Multidimensional Separation Techniques for Mitochondrial Proteomics

Dec 22, 2005 - ... Meisinger,*,‡ and Albert Sickmann*,†. Protein Mass Spectrometry and Functional Proteomics Group, Rudolf-Virchow-Center for Expe...
0 downloads 0 Views 335KB Size
Download PDF
Toward the Complete Yeast Mitochondrial Proteome: Multidimensional Separation Techniques for Mitochondrial Proteomics Joerg Reinders,† Rene´ P. Zahedi,† Nikolaus Pfanner,‡ Chris Meisinger,*,‡ and Albert Sickmann*,† Protein Mass Spectrometry and Functional Proteomics Group, Rudolf-Virchow-Center for Experimental Biomedicine, Julius-Maximilians-Universita¨t Wu ¨ rzburg, Versbacher Straβe 9, 97078 Wu ¨ rzburg, Germany, Institut fu ¨ r Biochemie und Molekularbiologie, Universita¨t Freiburg, Hermann-Herder-Straβe 7, 79104 Freiburg, Germany Received December 22, 2005

Proteomic analyses of different subcellular compartments, so-called organellar proteomics, facilitate the understanding of cellular functions on a molecular level. In this work, various orthogonal multidimensional separation techniques both on the protein and on the peptide level are compared with regard to the number of identified proteins as well as the classes of proteins accessible by the respective methodology. The most complete overview was achieved by a combination of such orthogonal techniques as shown by the analysis of the yeast mitochondrial proteome. A total of 851 different proteins (PROMITO dataset) were identified by use of multidimensional LC-MS/MS, 1D-SDSPAGE combined with nano-LC-MS/MS and 2D-PAGE with subsequent MALDI-mass fingerprinting. Our PROMITO approach identified the 749 proteins, which were found in the largest previous study on the yeast mitochondrial proteome, and additionally 102 proteins including 42 open reading frames with unknown function, providing the basis for a more detailed elucidation of mitochondrial processes. Comparison of the different approaches emphasizes a bias of 2D-PAGE against proteins with very high isoelectric points as well as large and hydrophobic proteins, which can be accessed more appropriately by the other methods. While 2D-PAGE has advantages in the possible separation of protein isoforms and quantitative differential profiling, 1D-SDS-PAGE with nano-LC-MS/MS and multidimensional LCMS/MS are better suited for efficient protein identification as they are less biased against distinct classes of proteins. Thus, comprehensive proteome analyses can only be realized by a combination of such orthogonal approaches, leading to the largest dataset available for the mitochondrial proteome of yeast. Keywords: Saccharomyces cerevisiae • mitochondria • proteomics • PROMITO • multidimensional separation

Introduction Proteome analysis of entire cell organelles has been facilitated by the enormous progress concerning separation and detection techniques in recent years. Particularly multidimensional separations coupled to tandem mass spectrometry have gained broad popularity. Thereby, also difficult-to-access proteins such as low-abundant or membrane proteins have become amenable, granting deeper insight into the cellular network and its functions. Prefractionation of cellular components has proven to be a prerequisite of high importance.1 The advantages of subcellular fractionation consist in both reduction of sample complexity, granting access to lowabundant proteins, and gaining additional information concerning protein localization. Various techniques, such as * To whom correspondence should be addressed. (A.S.) Tel: +49 931 201 48730. Fax: +49 931 201 48123. E-mail: albert.sickmann@ virchow.uni-wuerzburg.de. (C.M.) Tel: +49 761 203 5287. Fax: +49 761 203 5261. E-mail: [email protected]. † Julius-Maximilians-Universita¨t Wu ¨ rzburg. ‡ Universita¨t Freiburg. 10.1021/pr050477f CCC: $33.50

 2006 American Chemical Society

gradient centrifugation, free-flow-electrophoresis, and immunoprecipitation have been used to isolate distinct cell compartments.2-9 The purity obtained by such isolation procedures is crucial for subsequent proteome analyses to avoid false positives and thus has to be monitored by suitable methods, e.g., Western-blotting or electron microscopy. Generally, entire organelle proteomes are too complex to be sufficiently separated by one-dimensional methods. Therefore, multidimensional protein and peptide separation techniques coupled to MS- and MS/MS-systems have been used for in depth analysis of organellar proteomes1,9-11 such as 2D-PAGE with subsequent MALDI-mass fingerprinting or shotgun approaches. Different approaches toward an inventory of the mitochondrial proteome9,12-20 have facilitated analyses on a molecular level. Mapping of (sub-) proteomes by 2D-PAGE and subsequent MALDI-mass fingerprinting has a long-standing record,21-25 but the inherent limitations of this approach require further orthogonal methods for the separation task at hand.26 Both 1D-PAGE with subsequent LC-MS/MS and MDLCapproaches have been used for unbiased analyses of such Journal of Proteome Research 2006, 5, 1543-1554

1543

Published on Web 06/06/2006

research articles

Figure 1. (a) Workflow scheme for the applied methods of the different multidimensional separation techniques. The separation task is shifted from the protein to the peptide level (2D-PAGE: two protein separation dimensions, 1D-PAGE/nano-LC: onedimensional protein and one-dimensional peptide separation, 2DLC: two separation dimensions on the peptide level). Thereby, different proteins can be accessed leading to a more complete overview of the addressed (sub-) proteome. (b) Number of different proteins identified by the respective methods. Only 137 proteins were detected by all methods, pointing out the necessity for combination of the various approaches.

highly complex samples (see Figure 1a). A combination of these orthogonal proteomics techniques is critical for the generation of proteome maps as complete as possible. In this work, we investigated the advantages and disadvantages of the different methods upon the generated dataset (PROMITO), composed of 851 mitochondrial proteins of the yeast Saccharomyces cerevisiae, representing a sufficiently large dataset for a comprehensive comparison.

Materials and Methods Isolation of Mitochondria from Saccharomyces cerevisiae. Yeast cells (strain YPH499) were grown to an OD of 1.5-2.0 in YPG medium (1% yeast extract, 2% bactopeptone, 3% glycerol, pH 5.0). Isolation of mitochondria was accomplished according to Meisinger et al.27 Briefly, a crude mitochondrial fraction was obtained by differential centrifugation and adjusted to a protein concentration of 5 mg/mL in SEM-buffer (250 mM sucrose, 1 mM EDTA, 10 mM MOPS, pH 7.2). Mitochondria were treated by 10 strokes with a glass-Teflon potter and loaded onto a 1544

Journal of Proteome Research • Vol. 5, No. 7, 2006

Reinders et al.

three-step sucrose gradient (1.5 mL 60%, 4 mL 32%, 1.5 mL 23%, 1,5 mL 15% sucrose in EM-buffer (1 mM EDTA, 10 mM MOPS, pH 7.2)) and centrifuged for 1 h at 134 000 × g, yielding highly pure mitochondria at the 60/32%-sucrose-interface. The sucrose gradient step was performed twice. Purity of the obtained mitochondria was checked by Western-blot using antisera against various cellular compartments9,27,28 (see Supporting Information Figure 1). Two-Dimensional Gel Electrophoresis with Subsequent MALDI Mass Fingerprinting. Highly pure mitochondria were lysed by ultrasonication in sample buffer (7 M urea, 2 M thiourea, 2% CHAPS, 0.002% bromophenolblue) containing three glass beads (1 mm diameter). 2D-PAGE was carried out with at least three replicates as described previously29,30 using pH 3-10-NL-IPG-strips (Amersham Biosciences, Freiburg, Germany) and self-casted 12.5%-gels for the second dimension. The gels were either silver-stained according to Blum et al.31 or stained by colloidal Coomassie.32 Protein spots were excised, washed 3x alternately with 50 mM NH4HCO3, pH 7.8, and 25 mM NH4HCO3, pH 7.8, 50% acetonitrile, digested tryptically and subjected to MALDI-mass fingerprinting using an Ultraflex TOF/TOF (Bruker Daltonics, Bremen, Germany). If no positive identification could be obtained by database searching using the Mascot-algorithm (version 2.0)33 against SGD (Saccharomyces Genome Database, http://www.yeastgenome.org/),34 then LIFT-MS/MS-spectra of distinct peptide ion signals were recorded and also searched against SGD. SDS-PAGE and Subsequent nano-LC-MS/MS. Purified mitochondria (100 µg protein) were lysed in LDS-sample buffer (106 mM Tris-HCl, 141 mM Tris base, 2% (w/v) LDS (lithium dodecyl sulfate), 10% (v/v) glycerol, 0.51 mM EDTA, 0.22 mM SERVA Blue G250, 0.175 mM Phenol Red, pH 8.5) upon addition of three glass beads by 3× alternate ultrasonication/incubation steps on ice. SDS-PAGE and subsequent nano-LC-MS/MS were performed according to Wagner et al.35 The obtained MS/ MS-spectra were searched against SGD using Mascot-33 and Sequest-algorithms (version 2.7).36 The experiment was repeated three times. Multidimensional Liquid Chromatography (MDLC). Mitochondria were subjected to proteolytic digest using 4 different proteases (trypsin, chymotrypsin, Glu-C, subtilisin): purified mitochondria (2 mg protein) were suspended in water, lysed by sonification with 3× alternate ultrasonication/incubation steps on ice and digested using 50 µg of the respective protease in 100 mM NH4HCO3, pH 7.8, overnight or in case of subtilisin for 2 h at 37°C. Aliquots of 50 µg were used for 1D-RP-LCMS/MS; the residual sample volumes were divided in half and subjected to MDLC using either offline-SAX-RP-LC-MS/MS or offline-SCX-RP-MS/MS according to Wagner et al.35 The obtained MS/MS-spectra were searched against SGD using Mascot-33 and Sequest-algorithms.36 The experiments were performed twice for each protease. Tandem Mass Spectrometry, Database Searching and Interpretation of Results. For the database searches of MALDIfingerprints and LIFT-spectra the following parameters were used: trypsin as protease, one missed cleavage, cysteines carbamidomethylated (fixed), methionines oxidized (variable), peptide tolerance 0.1 Da, fragment ion tolerance 0.5 Da (for LIFT-spectra). Nano-LC-MS/MS-experiments were performed using a linear ion trap (QTrap4000, Applied Biosystems, Framingham, MA) applying a spray voltage of 2200 V or with a 3D-ion trap (LCQ Deca XPplus, Thermo Electron, Dreieich, Germany) and a

Toward the Complete Mitochondrial Proteome

spray voltage of 2000 V. In both cases a survey scan (350-2000 Da for linear IT/400-2000 Da for 3D-IT) was followed by three MS/MS scans of the most intensive signals. Database search parameters for the ESI-MS/MS-spectra were set differently depending on the mass spectrometer. For linear ion trap spectra doubly and triply charged ions were considered and a mass deviance of 0.2 Da was used while for 3D-ion trap spectra a mass deviance of 1.5 Da was applied for singly, doubly and triply charged ions. For Sequest-results spectra with ∆CN > 0.1 and xCorr > 1.8, 2.5, 3.5 for singly, doubly and triply charged ions37 and in case of Mascot-results only spectra with p < 0.01 were selected regardless of the used protease. These spectra were subjected to manual verification checking for completeness of ion series, signal-to-noise-ratio, nonmatched signals and sequencespecific fragmentation features, e.g., proline-breaches. Comparison of the PROMITO Dataset to Previous Studies on the Yeast Mitochondrial Proteome. The entire dataset obtained from the different multidimensional analysis approaches has been compared to the mitochondrial protein reference set of the MitoP2-database38 (version 2-2004) and the various studies on the yeast mitochondrial proteome referenced therein.9,12,39,40 Manual inspection of the reference set revealed some proteins for which a mitochondrial localization was just predicted or data were derived from other highthroughput studies. Furthermore, it contains 10 intron-derived potential gene products from the mitochondrial genome, which were not confirmed as stable products. Thus, the following 16 protein entries were subtracted from the reference dataset: Q0075, Q0160, Q0065, Q0060, Q0110, Q0070, Q0120, Q0055, Q0115, Q0050, YIL006w, YGR008c, YLR165c, YER058w, YDL185w, and YNL292w. Comparison of the different proteome studies with the MitoP2-database was done using both the entire and the corrected dataset.

Results and Discussion Highly purified yeast mitochondria were subjected to different separation and analysis methods outlined in Figure 1a. Separation by 2D-PAGE (see Supporting Information Figure 2) and analysis by MALDI-MS led to the identification of 169 different proteins (see Supporting Information Table 1) while analysis by SDS-PAGE and subsequent nano-LC-MS/MS resulted in 630 different identifications (see Supporting Information Table 2; Figures 1a,b and 2a). The multidimensional liquid chromatography approach yielded 491 proteins altogether (see Supporting Information Table 3) with less than 50% overlap between the resulting protein identifications of SCX/RP- and SAX/RP-LC. Furthermore, the redundancy of the MDLCdatasets was ∼59.5%. MALDI mass fingerprints were verified by an additional LIFT-MS/MS-spectrum per identified protein, for MS/MS-analyses at least two different peptides identified by significantly scored and manually validated spectra were required for positive identification. In comparison to our previous study,9 which had been the most extensive study on the yeast mitochondrial proteome to date with 749 different proteins identified, 102 additional proteins were identified (see Table 1 and supplementary Table 4), showing the enhanced sensitivity and resolution of the recent work (>2/3 of new proteins have codon adaptation indices (CAI) < 0.1541). Furthermore, a comparison of the two datasets with regard to protein copy numbers (Yeast GFP Database; http:// yeastgfp.ucsf.edu/) also indicated a higher sensitivity of the recent study as the set of newly identified proteins features a

research articles

Figure 2. (a) Number of different proteins identified in total and by the individual multidimensional separation approaches. 2DPAGE yielded the lowest number of identifications while 1DPAGE/nano-LC-MS/MS provided the best results. b) Lowabundant proteins according to codon adaptation index (CAI < 0.15). With 2D-PAGE the lowest percentage of proteins of lowabundance was obtained while both 1D-PAGE/nano-LC-MS/MS and 2D-LC-MS/MS showed higher rates. The total number of different proteins identified by the respective method(s) was set to 100%.

higher degree of low-abundant proteins (see Supporting Information Figure 3). Thus, we identified considerably more proteins with low copy number per cell, e.g., the remaining constituents of the mitochondrial protein transport machinery (Sam37, Tim8, Tim9) or regulators of mitochondrial morphology (Mdm34. Mdv1). Comparison of the results from the distinct multidimensional analyses (2D-PAGE-, SDS-PAGE/nano-LC-MS/MS- and 2DLC-MS/MS-based approaches) yielded just five proteins identified solely by 2D-PAGE, whereas 338 proteins were uniquely detected by SDS-PAGE/nano-LC-MS/MS, and 206 proteins only detected using 2D-LC-MS/MS (see Figure 1b/Supporting Information Table 5). Only 137 proteins (16.1%) were identified by each of these methods, indicating that only a combination of these orthogonal methods is able to grant an unbiased overview of the mitochondrial proteome. Examination of the physicochemical properties of the identified proteins with regard to the respective analysis method shows a bias of our 2D-PAGE-based approach against certain classes of proteins. Low-abundant proteins are strongly underrepresented in our 2D-PAGE compared to the other methods (see Figure 2b), as assessed by the CAI (10, only 11.2% of the proteins detected on the 2D-gels are within this region (see Figure 3a and d). However, the rate of proteins with pI > 10 identified by the other methods is above 20% (27.9% for SDSPAGE/nano-LC-MS/MS and 22.8% for the 2D-LC-MS/MSapproach) (see Figure 3b-d). Large proteins (MW > 100 kDa/ 14.8% of all identified proteins) are also better accessible by 1550

Journal of Proteome Research • Vol. 5, No. 7, 2006

mammalian homologsc

c

According YPD

SDS-PAGE/nano-LC-MS/MS (11.6% of the identified proteins) and 2D-LC-MS/MS (13.6%) than by 2D-PAGE (1.2%/see Figure 3a-c and e). This comparison shows also advantages of the SDS-PAGE/nano-LC-MS/MS-approach over 2D-LC-MS/MS for highly basic proteins (see Figure 3d). The bias of our 2D-PAGE against hydrophobic and transmembrane proteins is shown in Figure 4 with regard to the GRAVY index43 and number of predicted TMDs according to the SGD-database. Both SDS-PAGE/nano-LC-MS/MS- (10.8% GRAVY > 0 and 22.5% more than one TMD) and 2D-LC-MS/ MS-based (12.2% GRAVY > 0 and 26.1% more than one TMD) approaches provide better results than 2D-PAGE (3.6% GRAVY > 0 and 5.9% more than one TMD). With the identification of 851 different proteins we have generated the most comprehensive dataset for the yeast mitochondrial proteome (PROMITO, see supplementary Table 6). In addition to 749 proteins identified in our previous study,9 we present 102 newly identified proteins including 42 open reading frames with unknown function.34 Seventy-three of these proteins have homologues in human, rat or mouse (Table 1) indicating the importance of analyzing subcellular proteoms in model organisms such as yeast. These data could further lead to identification of novel human disease genes involved in mitochondrial disorders. Indeed one of the new identified open reading frames, Ypr125w (now termed Ylh47) has homology with the human LETM1 protein which is involved in the

research articles

Toward the Complete Mitochondrial Proteome

Figure 3. Identified proteins plotted according to pI and molecular weight. The entirety of all proteins is given in green; the proteins identified by the respective method are shown in red: (a) 2D-PAGE-approach, (b) 1D-PAGE/nano-LC-MS/MS, (c) 2D-LC-MS/MS. The 2D-PAGE-approach was biased against proteins with high molecular weights. The cluster of proteins with pI > 10 was also underrepresented. Both 1D-PAGE/nano-LC-MS/MS and 2D-LC-MS/MS yielded less biased results. (d) Highly basic proteins (pI > 10). Although 26.0% of all identified proteins belonged to this group the proportion detected by the 2D-PAGE-approach was only 11.2%. The proportions detected by each the 1D-PAGE/nano-LC-MS/MS- and the 2D-LC-MS/MS-based methods were above 20%. The highest rate of identification of basic proteins was obtained with the 1D-PAGE/nano-LC-MS/MS-approach. The total number of different proteins identified by the respective method(s) was set to 100%. (e) Proteins with high molecular weight (MW > 100 kDa). These proteins were strongly underrepresented in the 2D-PAGE approach. Even though the rate was higher for the 1D-PAGE/nano-LC-MS/MS-approach, the highest rate was achieved by the 2D-LC-MS/MS technique. The total number of different proteins identified by the respective method(s) was set to 100%.

Wolf-Hirschhorn Syndrome.44,48 The dataset presented here covers 81% of the actual mitochondrial reference set (521 proteins) listed in the MitoP2 database12 (see Table 2). This reference set was generated according to published experimental data and by manual annotation.12 We noticed, however, that this set lists some nonconfirmed mitochondrial proteins, for which a mitochondrial localization was only predicted or derived from other high-throughput studies, like the GFPdatabase.45 It also contains 10 intron-derived potential gene products from the mitochondrial genome, which were not confirmed as stable products so far. We thus generated a corrected reference set with 505 proteins. The coverage of the PROMITO dataset can then be calculated as 84% (see Table 2). By using the MitoP2 reference set, our previous mitochondrial proteome9 yields a coverage of 74% or 76% with the modified reference set. Other proteome studies on mitochondrial proteins achieved coverages of 51/53%,12 30/31%,40 and 27/28%39 (see Table 2).

Conclusion Evaluation of the analysis of the yeast mitochondrial proteome with regard to the applied separation techniques reveals that the used 2D-LC-MS/MS-protocol yields good rates for the detection of hydrophobic and membranous as well as very large proteins. However, SDS-PAGE with subsequent nano-LC-MS/ MS shows better identification rates for highly basic proteins and, most importantly, for the total number of different proteins identified. The 2D-PAGE-based approach reveals biases against certain classes of proteins such as proteins of high pI, large size or hydrophobic nature. Furthermore, it yielded the lowest number of protein identifications which could at least partially be overcome by the use of narrow-range pH-strips for isoelectric focusing and gradient gels in the second dimension for better resolution of high molecular weight proteins. The number of identifiable proteins could further be enhanced by application of additional staining procedures.42 Moreover, the fundamental advantages of this Journal of Proteome Research • Vol. 5, No. 7, 2006 1551

research articles

Reinders et al.

Figure 4. Identified proteins plotted according to hydropathicity (GRAVY index) and number of predicted transmembrane domains (TMDs). The entirety of all proteins is given in green, the proteins identified by the respective method are shown in red: (a) 2D-PAGEapproach, (b) 1D-PAGE/nano-LC-MS/MS, (c) 2D-LC-MS/MS. Very hydrophobic proteins, e.g., with many membrane-spanning domains, are not suitable for 2D-PAGE-based methods. The other two methods yielded less biased results. 2D-LC-MS/MS showed the highest identification rates for both proteins with a positive GRAVY index (overall hydrophobicity) (d) and proteins with multiple predicted TMDs (e). The total number of different proteins identified by the respective method(s) was set to 100%.

method are high reproducibility, quantifiability and the ability to resolve protein isoforms not only with regard to size but also pI. Therefore, this technique bears unique properties for the differential analysis of complex samples in comparison to the other methods. The accessibility of low-abundant proteins is enhanced by the use of the SDS-PAGE/nano-LC-MS/MS and 2D-LC-MS/MS rather than 2D-PAGE. The analysis of the yeast mitochondrial proteome by orthogonal multidimensional approaches led to the detection of 851 different proteins (PROMITO dataset). 102 additional identifications have been achieved compared to our previous study9 presumably due to enhanced sensitivity and resolution. The PROMITO dataset covers about 84% of the currently known mitochondrial proteome and thus represents the most comprehensive study so far. Abbreviations. 1D, one-dimensional; 2D, two-dimensional; CAI, codon adaptation index; CHAPS, 3-(3-cholamidopropyl)diethyl-ammonio-1-propanesulfonate; EDTA, ethylenediaminetetraacetic acid; ESI, electrospray ionization; GRAVY, grand average of hydropathicity; LC, liquid chromatography; LDS, lithium dodecyl sulfate; MALDI, matrix-assisted laser desorption/ionization; MDLC, multidimensional liquid chromatography; MOPS, 3-(N-morpholino)propanesulfonic acid; MS, mass spectrometry; MS/MS, tandem mass spectrometry; OD, optical density; ORF, open reading frame; PAGE, polyacryla1552

Journal of Proteome Research • Vol. 5, No. 7, 2006

mide gelelectrophoresis; pI ) isoelectric point; RP, reversed phase; SAX, strong anion exchange; SCX, strong cation exchange; SDS, sodium dodecyl sulfate; SGD, Saccharomyces genome database; TMD, transmembrane domain; YPD, yeast proteome database.

Acknowledgment. We thank Drs. S. Rospert, T. Sommer, W. Kunau, and T. H. Stevens for antisera. This work was supported by the Deutsche Forschungsgemeinschaft (Si 835/ 2-1,2 and Me1921/1-1,2), FZT-82 (DFG-Forschungszentrum), Sonderforschungsbereich 388, Gottfried Wilhelm Leibniz Program, Max Planck Research Award and the Fonds der chemischen Industrie. Supporting Information Available: Purity of the obtained mitochondria checked by Western-blot (SI Figure 1). Separation of mitochondrial proteins by 2D-PAGE (SI Figure 2). Comparison of copy numbers of newly identified proteins (SI Figure 3). 169 different proteins identified by 2D-PAGE (SI Table 1). 630 different identifications by 1D-PAGE (SI Table 2), and 491 proteins yielded by the multidimensional liquid chromatography approach (SI Table 3). The peptides identified from Table 1 (SI Table 4), list of all proteins identified by the multidimensional separation methods (SI Table 5), and list of all proteins of the PROMITO dataset (SI Table 6). This material is available free of charge via the Internet at http://pubs.acs.org.

research articles

Toward the Complete Mitochondrial Proteome

References (1) Stasyk, T.; Huber, L. A. Zooming in: fractionation strategies in proteomics. Proteomics 2004, 4 (12), 3704-3716. (2) Bauer, A.; Kuster, B. Affinity purification-mass spectrometry. Powerful tools for the characterization of protein complexes. Eur. J. Biochem. 2003, 270 (4), 570-578. (3) Dreger, M. Subcellular proteomics. Mass Spectrom. Rev. 2003, 22 (1), 27-56. (4) Dreger, M. Proteome analysis at the level of subcellular structures. Eur. J. Biochem. 2003, 270 (4), 589-599. (5) Fialka, I.; Pasquali, C.; Lottspeich, F.; Ahorn, H.; Huber, L. A. Subcellular fractionation of polarized epithelial cells and identification of organelle-specific proteins by two-dimensional gel electrophoresis. Electrophoresis 1997, 18 (14), 2582-2590. (6) Hannig, K.; Heidrich, H. G. The use of continuous preparative free-flow electrophoresis for dissociating cell fractions and isolation of membranous components. Methods Enzymol. 1974, 31 (Pt A), 746-761. (7) Pasquali, C.; Fialka, I.; Huber, L. A. Subcellular fractionation, electromigration analysis and mapping of organelles. J. Chromatogr. B Biomed. Sci. Appl. 1999, 722 (1-2), 89-102. (8) Reinders, J.; Lewandrowski, U.; Moebius, J.; Wagner, Y.; Sickmann, A. Challenges in mass spectrometry-based proteomics. Proteomics 2004, 4 (12), 3686-3703. (9) Sickmann, A.; Reinders, J.; Wagner, Y.; Joppich, C.; Zahedi, R.; Meyer, H. E.; Schonfisch, B.; Perschil, I.; Chacinska, A.; Guiard, B.; Rehling, P.; Pfanner, N.; Meisinger, C. The proteome of Saccharomyces cerevisiae mitochondria. Proc. Nat. Acad. Sci. 2003, 100 (23), 13207-13212. (10) Taylor, S. W.; Fahy, E.; Ghosh, S. S. Global organellar proteomics. Trends Biotechnol. 2003, 21 (2), 82-88. (11) Huber, L. A.; Pfaller, K.; Vietor, I. Organelle proteomics: implications for subcellular fractionation in proteomics. Circ. Res. 2003, 92 (9), 962-968. (12) Prokisch, H.; Scharfe, C.; Camp, D. G. 2nd; Xiao, W.; David, L.; Andreoli, C.; Monroe, M. E.; Moore, R. J.; Gritsenko, M. A.; Kozany, C.; Hixson, K. K.; Mottaz, H. M.; Zischka, H.; Ueffing, M.; Herman, Z. S.; Davis, R. W.; Meitinger, T.; Oefner, P. J.; Smith, R. D.; Steinmetz, L. M. Integrative analysis of the mitochondrial proteome in yeast. PLoS Biol. 2004, 2 (6), e160. (13) Rezaul, K.; Wu, L.; Mayya, V.; Hwang, S. I.; Han, D. A systematic characterization of mitochondrial proteome from human T leukemia cells. Mol. Cell. Proteomics 2005, 4 (2), 169-181. (14) Millar, A. H.; Heazlewood, J. L.; Kristensen, B. K.; Braun, H. P.; Moller, I. M. The plant mitochondrial proteome. Trends Plant Sci. 2005, 10 (1), 36-43. (15) Gabaldon, T.; Huynen, M. A. Shaping the mitochondrial proteome. Biochim. Biophys. Acta 2004, 1659 (2-3), 212-220. (16) Gibson, B. W. The human mitochondrial proteome: oxidative stress, protein modifications and oxidative phosphorylation. Int. J. Biochem. Cell Biol. 2005, 37 (5), 927-934. (17) Reichert, A. S.; Neupert, W. Mitochondriomics or what makes us breathe. Trends Genet. 2004, 20 (11), 555-562. (18) Gaucher, S. P.; Taylor, S. W.; Fahy, E.; Zhang, B.; Warnock, D. E.; Ghosh, S. S.; Gibson, B. W. Expanded coverage of the human heart mitochondrial proteome using multidimensional liquid chromatography coupled with tandem mass spectrometry. J. Proteome Res. 2004, 3 (3), 495-505. (19) Taylor, S. W.; Warnock, D. E.; Glenn, G. M.; Zhang, B.; Fahy, E.; Gaucher, S. P.; Capaldi, R. A.; Gibson, B. W.; Ghosh, S. S. An alternative strategy to determine the mitochondrial proteome using sucrose gradient fractionation and 1D PAGE on highly purified human heart mitochondria. J. Proteome Res. 2002, 1 (5), 451-458. (20) Taylor, S. W.; Fahy, E.; Zhang, B.; Glenn, G. M.; Warnock, D. E.; Wiley: S.; Murphy, A. N.; Gaucher, S. P.; Capaldi, R. A.; Gibson, B. W.; Ghosh, S. S. Characterization of the human heart mitochondrial proteome. Nat. Biotechnol. 2003, 21 (3), 281-286. (21) Sutton, C. W.; Pemberton, K. S.; Cottrell, J. S.; Corbett, J. M.; Wheeler, C. H.; Dunn, M. J.; Pappin, D. J. Identification of myocardial proteins from two-dimensional gels by peptide mass fingerprinting. Electrophoresis 1995, 16 (3), 308-316. (22) Scheler, C.; Lamer, S.; Pan, Z.; Li, X. P.; Salnikow, J.; Jungblut, P. Peptide mass fingerprint sequence coverage from differently stained proteins on two-dimensional electrophoresis patterns by

(23)

(24)

(25)

(26) (27) (28) (29)

(30)

(31) (32)

(33)

(34)

(35) (36)

(37)

(38)

(39) (40) (41) (42)

(43)

matrix assisted laser desorption/ionization-mass spectrometry (MALDI-MS). Electrophoresis 1998, 19 (6), 918-927. Cohen, A. M.; Rumpel, K.; Coombs, G. H.; Wastling, J. M. Characterisation of global protein expression by two-dimensional electrophoresis and mass spectrometry: proteomics of Toxoplasma gondii. Int. J. Parasitol. 2002, 32 (1), 39-51. Watt, S. A.; Patschkowski, T.; Kalinowski, J.; Niehaus, K. Qualitative and quantitative proteomics by two-dimensional gel electrophoresis, peptide mass fingerprint and a chemically coded affinity tag (CCAT). J. Biotechnol. 2003, 106 (2-3), 287-300. Coelho, A.; de Oliveira Santos, E.; Faria, M. L.; de Carvalho, D. P.; Soares, M. R.; von Kruger, W. M.; Bisch, P. M. A proteome reference map for Vibrio cholerae El Tor. Proteomics 2004, 4 (5), 1491-1504. Santoni, V.; Molloy, M.; Rabilloud, T. Membrane proteins and proteomics: un amour impossible? Electrophoresis 2000, 21 (6), 1054-1070. Meisinger, C.; Sommer, T.; Pfanner, N. Purification of Saccharomcyes cerevisiae mitochondria devoid of microsomal and cytosolic contaminations. Anal. Biochem. 2000, 287 (2), 339-342. Meisinger, C.; Pfanner, N.; Truscott, K. N. Isolation of yeast mitochondria. Methods Mol. Biol. 2006, 313, 33-39. Sickmann, A.; Dormeyer, W.; Wortelkamp, S.; Woitalla, D.; Kuhn, W.; Meyer, H. E. Identification of proteins from human cerebrospinal fluid, separated by two-dimensional polyacrylamide gel electrophoresis. Electrophoresis 2000, 21 (13), 2721-2728. Sickmann, A.; Dormeyer, W.; Wortelkamp, S.; Woitalla, D.; Kuhn, W.; Meyer, H. E. Towards a high-resolution separation of human cerebrospinal fluid. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2002, 771 (1-2), 167-196. Blum, H.; Beier, H.; Gross, H. J. Improved Silver Staining of PlantProteins, Rna and DNA in Polyacrylamide Gels. Electrophoresis 1987, 8 (2), 93-99. Neuhoff, V.; Stamm, R.; Pardowitz, I.; Arold, N.; Ehrhardt, W.; Taube, D. Essential problems in quantification of proteins following colloidal staining with coomassie brilliant blue dyes in polyacrylamide gels, and their solution. Electrophoresis 1990, 11 (2), 101-117. Hirosawa, M.; Hoshida, M.; Ishikawa, M.; Toya, T. MASCOT: multiple alignment system for protein sequences based on threeway dynamic programming. Comput. Appl. Biosci. 1993, 9 (2), 161-167. Weng, S.; Dong, Q.; Balakrishnan, R.; Christie, K.; Costanzo, M.; Dolinski, K.; Dwight, S. S.; Engel, S.; Fisk, D. G.; Hong, E.; IsselTarver, L.; Sethuraman, A.; Theesfeld, C.; Andrada, R.; Binkley, G.; Lane, C.; Schroeder, M.; Botstein, D.; Michael Cherry, J. Saccharomyces Genome Database (SGD) provides biochemical and structural information for budding yeast proteins. Nucleic Acids Res. 2003, 31 (1), 216-218. Wagner, Y.; Sickmann, A.; Meyer, H. E.; Daum, G. Multidimensional nano-HPLC for analysis of protein complexes. J. Am. Soc. Mass Spectrom. 2003, 14 (9), 1003-1011. Eng, J. K.; McCormack, A. L.; Yates, I.; John R. An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J. Am. Soc. Mass Spectrom. 1994, 5 (11), 976-989. Tabb, D. L.; McDonald, W. H.; Yates, J. R., 3rd, DTASelect and Contrast: tools for assembling and comparing protein identifications from shotgun proteomics. J. Proteome Res. 2002, 1 (1), 2126. Prokisch, H.; Andreoli, C.; Ahting, U.; Heiss, K.; Ruepp, A.; Scharfe, C.; Meitinger, T. MitoP2: the mitochondrial proteome databases now including mouse data. Nucleic Acids Res. 2006, 34 (Database issue), D705-711. Pflieger, D.; Le Caer, J. P.; Lemaire, C.; Bernard, B. A.; Dujardin, G.; Rossier, J. Systematic identification of mitochondrial proteins by LC-MS/MS. Anal. Chem. 2002, 74 (10), 2400-2406. Ohlmeier, S.; Kastaniotis, A. J.; Hiltunen, J. K.; Bergmann, U. The yeast mitochondrial proteome, a study of fermentative and respiratory growth. J. Biol. Chem. 2004, 279 (6), 3956-3979. Sharp, P. M.; Li, W. H. The codon Adaptation Indexsa measure of directional synonymous codon usage bias, and its potential applications. Nucleic Acids Res. 1987, 15 (3), 1281-1295. Rabilloud, T.; Strub, J. M.; Luche, S.; van Dorsselaer, A.; Lunardi, J. A comparison between Sypro Ruby and ruthenium II tris (bathophenanthroline disulfonate) as fluorescent stains for protein detection in gels. Proteomics 2001, 1 (5), 699-704. Kyte, J.; Doolittle, R. F. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 1982, 157 (1), 105-132.

Journal of Proteome Research • Vol. 5, No. 7, 2006 1553

research articles (44) Hodges, P. E.; Payne, W. E.; Garrels, J. I. The Yeast Protein Database (YPD): a curated proteome database for Saccharomyces cerevisiae. Nucleic Acids Res. 1998, 26 (1), 68-72. (45) Huh, W. K.; Falvo, J. V.; Gerke, L. C.; Carroll, A. S.; Howson, R. W.; Weissman, J. S.; O’Shea, E. K. Global analysis of protein localization in budding yeast. Nature 2003, 425 (6959), 686-691. (46) Cherry, J. M.; Adler, C.; Ball, C.; Chervitz, S. A.; Dwight, S. S.; Hester, E. T.; Jia, Y.; Juvik, G.; Roe, T.; Schroeder, M.; Weng, S.; Botstein, D. SGD: Saccharomyces Genome Database. Nucleic Acids Res. 1998, 26 (1), 73-79.

1554

Journal of Proteome Research • Vol. 5, No. 7, 2006

Reinders et al. (47) Gasteiger, E.; Jung, E.; Bairoch, A. SWISS-PROT: connecting biomolecular knowledge via a protein database. Curr. Issues Mol. Biol. 2001, 3 (3), 47-55. (48) Frazier, A. E.; Taylor, R. D.; Mick, D. V.; Warscheid, B.; Stoepel, N.; Meyer, H. E.; Ryan, M. T.; Guiard, B.; Rehling, P. Mdm38 interacts with ribosomes and is a component of the mitochondrial protein export machinery. J. Cell. Biol. 2006, 172, 553564.

PR050477F