Prefractionation of Proteome by Liquid Isoelectric Focusing Prior to

Zhi-Bin Ning , Qing-Run Li , Jie Dai , Rong-Xia Li , Chia-Hui Shieh and Rong Zeng. Journal of Proteome Research 2008 7 (10), 4525-4537. Abstract | Ful...
0 downloads 0 Views 563KB Size
Download PDF
Prefractionation of Proteome by Liquid Isoelectric Focusing Prior to Two-Dimensional Liquid Chromatography Mass Spectrometric Identification Rong-Xia Li, Hu Zhou, Su-Jun Li, Quan-Hu Sheng, Qi-Chang Xia, and Rong Zeng* Research Center for Proteome Analysis, Key Laboratory of Proteomics, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 320 Yueyang Road, Shanghai 200031, China Received December 24, 2004

Due to the complexity of proteomes, developing methods of sample fractionation, separation, concentration, and detection have become urgent to the identification of large numbers of proteins, as well as the acquisition of those proteins in low abundance. In this work, liquid isoelectric focusing (LIEF) combined with 2D-LC-MS/MS was applied to the proteome of Saccharomyces cerevisiae. This yielded a total of 1795 proteins that were detected and identified by 30 fractions of protein prefractioantion. Categorization of these hits demonstrated the ability of this technology to detect and identify proteins rarely seen in proteome analysis without protein fractiontion. LIEF-2D-LC-MS/MS also produced improved resolution of low-abundance proteins. Furthermore, we analyzed the characteristics of proteins obtained by LIEF-2D-LC-MS/MS. 1103 proteins with CAI under 0.2 were identified, allowing us to specifically obtain detailed biochemical information on these kind proteins. It was observed that LIEF-2D-LC-MS/MS is useful for large-scale proteome analysis and may be specifically applied to systems with wide dynamic ranges. Keywords: large-scale proteome • fractionation • liquid isoelectric focusing (LIEF) • 2D-LC-MS/MS • low-abundance proteins

Introduction In recent years, there has been increased effort on the separation, quantification, and identification of all proteins in a cell or tissue.1 The difficulty of analyzing the proteome is that many low-abundance proteins are not detected due to a predicted dynamic range of up to 5 orders of magnitude.2 Therefore, the field of proteomics requires the development of effective fractionation and separation methods that offer high sensitivity, and wide dynamic range to detect more proteins in a cell or tissue.3 The most widely used method to separate proteins from cell lysates is two-dimensional polyacrylamide gel electrophoresis (2-D PAGE).4-8 In 2D-PAGE, proteins are separated in one dimension by isoelectric point (pI), and in the other by molecular weight (MW). This method is capable of resolving over 1000 proteins and providing a pattern of spots. The pattern of spots observed in the 2-D PAGE image is reproducible and is representative of the cell type being analyzed.9-11 2-D technology has a number of limitations that can be particularly troublesome in expression proteomics applications. For example, detecting low-abundance proteins can be problematic; some proteins may be missed simply due to their low copy numbers and the limitations in the amount of sample that can * To whom correspondence should be addressed. Phone: 86-21-54920170. Fax: 86-21-54920171. E-mail: [email protected].

1256

Journal of Proteome Research 2005, 4, 1256-1264

Published on Web 07/14/2005

be loaded onto gels. Comigration of proteins during separation can also mask the presence of low-abundance proteins. Highthroughput analysis of proteomes remains challenging because the individual extraction, digestion, and analysis of each spot from 2D-PAGE are tedious and time-consuming. Shotgun proteomics refers to the direct analysis of complex protein mixtures to rapidly generate a global profile of the protein components within the mixture. The protein mixture is proteolytically digested and the resulting peptides are further analyzed by LC-MS/MS analysis.12-21 This approach has been facilitated by the use of multidimensional protein identification technology (MudPIT), which employs 2-D liquid chromatography, consisting of ion exchange and reversed phase columns, followed by tandem mass spectrometry. MudPIT has shown promising results in analyzing the yeast proteome.22 However, using shotgun method for protein digests, more than thousands of peptides were separated by 2D-LC; peptides of highabundance proteins could be found in many steps of salt elution. The reason for this phenomenon may be because peptides from high abundance proteins share most of the digested peptide mixture. The high abundance peptides may distribute to different salt steps (known as overlapping). On the other hand, high abundance peptides may cover up the detection of the low abundance peptides. Therefore, the fractionation of digested peptides may not be enough to extract low-abundance peptides from high-abundance peptides. This 10.1021/pr049751g CCC: $30.25

 2005 American Chemical Society

Prefractionation of Proteome by LIEF and 2D-LC-MS/MS

indicates that it may be necessary for the extraction of lowabundance proteins from high-abundance proteins prior to peptide separation and identification. Many groups have used various prefractionation methods for improved 1-D or 2-D gels,23-26 including liquid IEF. Of all electrophoretic methods, isoelectric focusing offers the highest resolution and is best suited for preparative applications.27 Up to now, several commercial instruments for solution IEF have been developed, including the Rotofor, RF3 (Recycling Free Flow Focusing protein fractionator), the Isoprime, and the IsoelectrIQ2.9,27-40,42 In this work, we have developed a method to separate proteins prior to protein digestion and 2D-LC-MS/ MS identification according to their pI, using LIEF in the RF3. Protein pre-fractionation diluted the high abundance proteins by distributing different forms of high abundance proteins into several fractions; the concentration of low abundance proteins could also be increased in the fractionated area, allowing easier detection. This report demonstrates that a large number of proteins, especially low-abundance proteins, can be separated using LIEF. A theoretical 2-D image of these proteins can be generated for the purpose of observing distinctive patterns. Automation and the speed of analysis are also greatly facilitated because the proteins remain in the liquid phase throughout the separation. LIEF-2D-LC-MS/MS is shown to be an alternative technique for large scale identification of proteomes as well as for the detection and enrichment of low-abundance proteins. Because the system is largely unbiased, proteins from all sub cellular portions of the cell with extremes in pI, MW, abundance, and hydrophobic characteristics were identified.

Experimental Section Yeast Sample Preparation. The yeast strain used in this study is from the YPH499 family. Cells were grown in 1% yeast extract/2% peptone (YP) t 2% glucose (YPD) medium. YPD499 was grown to mid-log phase (OD 0.6) in YPD at 30 °C and centrifuged to collect the pallets of the yeast cell. The pallets were washed using ddH2O and vortexed for 2 min to suspend the pallets, then centrifuged at 6000 rpm at 4 °C for 5 min, then the supernatant was discarded. The yeast cell pellets were then suspended in 1% n-octyl R-Dgalactopyranoside (OG) (Sigma), 1% Biolyte ampholytes, pH 3.5-10 (Amersham Biosciences), 10% Glycerol (Amersham Biosciences), 7 M urea (Bio-Rad), 2 M thiourea (ICN), 1% NP40 (USB), 1 Triton X-100 (USB), 10 mM DTT (Bio-Rad); and 1 mM PMSF, and vortexed for 2 min to effect cell disruption and protein solubilization.9 The suspension solution was lysated using a high-pressure homogenizer (EmulsiFlex1-C5) to rupture the yeast cells.43 The whole cell protein extract was then diluted to 105 mL with the LIEF buffer, and introduced into the RF3 separation chamber. LIEF Separation. A preparative-scale RF3 (Rainin) was used during separation and fractionation. This device separated proteins in the liquid-phase according to their pI. A normal load is capable of at least 20 to 40 mg protein per run, and as much as 1 to 2 g of protein with a total buffer volume of 105 mL. Approximately 20 mg of protein from the whole cell lysate was diluted to a final volume of 105 mL; the separation buffer contained 1% ampholytes (pH 3.5-10, Amersham Biosciences), 10% Glycerol, 8 M urea, 1% NP-40, 1 Triton X-100, 1 OG, 10 mM DTT, and 1 mM PMSF. These proteins were separated by isoelectric focusing over a 12-h period at a separation temperature of 10 °C. The procedure used for running the LIEF was a

research articles modified version of the standard procedure described in the manual from Rainin. The 30 fractions contained in the LIEF were collected simultaneously into the focusing channel. Then, using a peristaltic pump, pumped the separation fraction to the 30 tube fraction collector. A three-milliliter portion of each LIEF fraction was allocated into polypropylene micro-centrifuge tubes and could be stored at -80 °C for further analysis if necessary. The concentration of protein in each fraction was determined via the Bio-Rad Bradford-based protein assay. The pH of the fractions was determined using pH apparatus (Beckman). SDS-PAGE Separation. Yeast proteins resolved by LIEF separation into discrete pI ranges were further resolved according to their apparent molecular weight, by SDS-PAGE. Samples of LIEF fractions were suspended in an equal volume (6 µL) of sample buffer (125 mM Tris (pH 6.8) containing 1% SDS, 10% glycerol, 1% dithioerythritol (DTE), and bromophenol blue), and boiled for 15 min. They were then loaded onto 12% acrylamide gels. First, the samples underwent electrophoresis at 15 mA for 1 h, and then 20 mA until the dye front reached the opposite end of the gel. The resolved proteins were visualized by the method of modified silver staining.44 In abbreviation, the gels were fixed 1 h in 40% ethanol containing 10% acetic acid and overnight in 5% ethanol containing 5% acetic acid then washed successively (for 20 min each) in ddH2O. The gels were stained by modified silver staining method. Two-Dimensional Gel Electrophoresis. LIEF and the seconddimensional SDS-PAGE. 2-D gel electrophoresis was performed essentially as reported5,44-46 and the manufacturer’s instructions. LIEF was carried out at a maximum current setting of 70 µA/strips using a commercially available, dedicated apparatus: IPGphor (Amersham Pharmacia Biotech). Samples of 30µg protein for analytical gels were applied on pH 3-10 L IPG strips in a holder at their acidic ends. Samples were diluted to 125 µL with a rehydration buffer (8 M Urea, 2% CHAPS, 15 mM DTT and 0.5% v/v pH 3-10 L IPG buffer, trace bromophenol blue) and applied to strips by the following procedure: (1) 30 V, 20 °C step-n-hold, 12 h; (2) 500 V, 20 °C, gradient, 30 min; (3) 1000 V, 20 °C, gradient, 30 min; (4) 5000 V, 20 °C, stepn-hold, 10 h. After the first dimensional LIEF, IPG gel strips were placed in an equilibration solution (6 M Urea, 2% SDS, 30% glycerol, 50 mM Tris-HCl, pH 8.8) containing 10 mg/mL DTT for 15 min with shaking at 50 rpm on an orbital shaker. The gel strips were then transferred to the equilibration solution containing 25 mg/mL IAA and shaken for a further 15 min before being placed on a 12.5% SDS-polyacrylamide gel slab. The separation was carried out at a current setting of 15 mA/ gel for the initial 1 h, and 25 mA/gel thereafter. The seconddimensional SDS-PAGE was developed until the bromophenol blue dye marker had reached the middle of the gel. Protein Visualization and Image Analysis. Following the second-dimensional SDS-PAGE, silver staining was performed according to published methods.44,45 Protein patterns in the gel were recorded as digitalized images using a high-resolution scanner (GS-710 Calibrated Imaging Densitometer, Bio-Rad). Removal of Ampholyte, Salt and Enzymatic Digestion. Protein purification was performed as described previously47-51 with a few modifications. First, the sample solution was mixed with 10% TCA (trichloroacetic acid) and then incubated 2 h on ice. The supernatant was carefully removed after being centrifuged at 14 000 rpm for 40 min at 4 °C. The dried pellets were dissolved in reduction buffer21,52-54 (6 M urea, 50 mM tris, Journal of Proteome Research • Vol. 4, No. 4, 2005 1257

research articles

Figure 1. pH value of RF3 30 fractions. A pH 11.23-4.28 gradient was formed.

pH 8.5). The mixture was diluted with reduction buffer to a total volume of 200 µL, and then mixed with 2 µL of 1M dithiothreitol (DTT). After that, the mixture was incubated at 37 °C for 2.5 h, and then alkalized with 10 µL of 100 mM IAA for 40 min in the dark, at room temperature. The protein mixtures were ultra-filtrated using a 3 K ultra filtration membrane into a 100 mM ammonium bicarbonate buffer with pH 8.5; the proteins were quantitated using the Bradford method, and 600 µg of proteins were incubated with trypsin (50:1 Promega) at 37 °C overnight. After digestion, the peptide mixtures were ultra-filtrated using a 10K ultra filtration membrane (microcon) to remove undigested proteins and trypsin. Finally, the ultrafiltrate was lyophilized. 2D-LC-MS/MS. Orthogonal 2D LC-MS/MS was performed using a ProteomeX Workstation (Thermo Finnigan, San Jose, CA).41 The system was fitted with a strong cation exchange column (SCX, 320 µm ID × 100 mm, DEV SCX, Thermo Hypersil-Keystone) and two C18 reversed phase columns (RP, 180 µm × 100 mm, BioBasic C18, 5 µm, Thermo HypersilKeystone). The salt steps used were 0, 25, 50, 75, 100, 150, 200, 400, and 800 mM NH4Cl synchronized with nine 140 min RP gradients. RP solvents were 0.1% formic acid in either water (A) or acetonitrile (B).The setting of the LCQ Deca Xplus iontrap mass spectrometer was the same as previous. SEQUEST Analysis. The SEQUEST algorithm was used to interpret MS/MS as previously18,19 described. The SEQUEST algorithm was run on each of the 9 data sets against the protein database for yeast proteins from the SWISS-PROT/TrEMBL proteome set for S. cerevisiae 10/2003 released, using the TurboSEQUEST program in the BioWorks 3.0 software suite. We used conservative criteria to determine the protein content of our. An accepted SEQUEST result had to have a ∆Cn score of at least 0.1 (regardless of charge state). Peptides identified by SEQUEST may have three different charge states (+1, +2, or +3), each of which results in a unique spectrum for the same peptide. Peptides with a +1 charge state were accepted only if they were fully tryptic and had a cross correlation (Xcorr) of at least 1.8. Peptides with a +2 charge state were accepted if they had an Xcorr g2.2. +3 peptides were accepted if they had an Xcorr g3.7. Single peptides that identified a protein alone were manually validated after meeting the above criteria.

Results and Discussion Sample Preparation for LIEF. For the RF3, each of the liquid electrophoresis runs is capable of being loaded with at least 20 mg proteins. 30 fractions can be collected after LIEF. 30 protein fractions ranging in pH from 4.28 to 11.23 (Figure 1) 1258

Journal of Proteome Research • Vol. 4, No. 4, 2005

Li et al.

were collected. We tested several kinds of detergent for the protein extraction and LIEF, including (1) 1% ampholytes (pH 3.5-10)10% Glycerol(2) 1% ampholytes (pH 3.5-10)10% Glycerol7 M urea, 2 M thiourea, 1% NP-40, 1 Triton X-100, 1 OG, 10 mM DTT, and 1 mM PMSF(3) 1% ampholytes (pH 3.5-10)10% Glycerol8 M urea, 1% NP-40, 1 Triton X-100, 1 OG, 10 mM DTT, and 1 mM PMSF(4) 1% ampholytes (pH 3.5-10)10% Glycerol1% OG, 8 M urea, 1% NP-40, 1 Triton X-100, 1 OG, 10 mM DTT, and 1 mM PMSF. When buffer (1) was used in LIEF, precipitation in fractions 19 to 21 was found during focusing. As for the buffer (2), Thiourea crystals came out in acidic areas and unavoidably interfered with protein solubility and focusing. Finally, we chose OG, a strong nonionic detergent. OG can greatly increase the solubility of proteins, especially the membrane proteins for our work, thus optimizing sample separation and focusing. Using this procedure, both cytosolic and membrane proteins were available for analysis. Validation of Protein Prefractionation after LIEF. Figure 2 displays the SDS-PAGE maps of the total lysate and the 30 fractions collected from the LIEF with equal volume loaded on the gel. The results of 1D SDS-PAGE shows that different fractions have different patterns, which indicates the distribution of proteins in different fractions, and the relative abundance of proteins of each fraction differed from the whole cell lysate. From fraction 1 to fraction 22, and fraction 24, along with the increase of the pH, high molecular weight proteins gradually decrease; consistent with the distributing of the proteins in the database. Figure 3 shows the 2DE map of total lysates, and fraction 6, 12, 20, 22, and 24. The amount of protein fractions with pH 5.0 to 6.0 is much higher than other fractions; this is validated by SDS-PAGE and the actual 2DE map (Figure 2 and Figure 3). However, it was observed that high abundance proteins had diffused into several neighboring fractions. The most concerning about LIEF is that the proteins cannot isoelectrically focus as effectively as it would in a gel; due to the diffusion of proteins in the solution. Lubman.D. et al.9 had reported that, when comparing the liquid and gel-phase images of R-enolase, it can be seen that in both cases, substantial spreading of the protein occurs over a wide pI range; this range spans from pI 6.5 to 9.5 in both the liquid and gel phase. For more acidic proteins such as actin, it appears that in the liquid phase the proteins are more scattered in the pI dimension than for the corresponding gel separated proteins9. It should also be noted that the pI of a given protein may vary significantly due to post-translational modifications such as phosphorylation and glycosylation. Protein Digestion after LIEF Fractionation. The LIEF separated fractions contains 1% w/v of ampholytes, which hindered the detection of proteins and peptides in the LC-MS-based analyses55,56. The function groups of carrier ampholytes may combine with the proteins and become difficult to remove. We tested many kinds of methods to remove the function groups, such as dialysis, precipitations, gel filtration, ion exchange, ultra filtration. We finally chose TCA precipitations plus ultra filtration. The TCA precipitations and ultra filtration cleanup of the LIEF fractions prior to LC-MS/MS analysis increases the capability of the analysis by eliminating the exposure of the reversed-phase column to urea and ampholytes. Identification of Protein by 2D-LC-MS/MS. 600 µg of protein was used in each fraction to identify proteins by 2DLC-MS/MS (data not shown). In total, 1795 proteins were identified with this strategy; 923 (51.42%) proteins were identi-

Prefractionation of Proteome by LIEF and 2D-LC-MS/MS

research articles

Figure 2. SDS-PAGE map of yeast total lysates and 30 fractions of RF3. An equal volume (6 µL) of each fraction was loaded and visualized by silver staining.

Figure 3. Two-dimensional electrophoresis of yeast proteins fractionated by the RF3 instrument. The actual 2-D map of the yeast total lysate (A). The 6 fractions are denoted as basic pI to acidic pI. Their pH range respectively is fraction 6 (B; pH 9.88-9.5), fraction 12 (C; pH 8.36-7.90), fraction 20 (D; pH 6.37-6.14), fraction 22 (E; pH), fraction 24 (F; pH 5.49-5.07).

fied based on at least 2 unique peptides. In contrast, we have identified 577 proteins in the yeast total lysate without fractionation before 2D-LC-MS/MS; proteins with minimum two unique peptides reach 266, covering 46.1%. By the LIEF-2DLC-MS/MS method, we can improve the number of identified proteins to 3.11-fold, compared to the 2D-LC-MS/MS. Moreover, the number of proteins with two unique identified peptides reached 3.45-fold over 2D-LC-MS/MS. The results indicated that the percentage of these more confident proteins was also increased. Generally, most of the digested peptides obtained from highabundance proteins are easily diffused and detected in all salt steps in 2D-LC-MS/MS, covering up the detection of peptides from low-abundance proteins. Therefore, it is more efficient to fractionate protein mixtures before peptide separation and detection. The clustering of high-abundance proteins into several limited fractions would result in the enrichment of lowabundant proteins. We introduced the LIEF method, which

clustered the most high-abundance proteins into several specified contiguous fractions while the low-abundance proteins could be concentrated in the other fractions. For example, the As a high-abundance protein, HS72_YEAST(CAI value 0.82) could be identified easily by capillary 2D-LC-ESI-MS/MS via introducing various derived peptides into the system. Although 9 SCX salt steps were used to fractionate the peptide mixtures from the total lysate of yeast, a number of peptides, even the same peptides from HS72_YEAST, were detected throughout of the 9 steps by MS. After the prefractionation of proteins from yeast, HS72_YEAST were clustered into limited fractions (e.g., fraction 25, 30, 29, 28, 27, 26, 24). Only in fraction 25 were peptides from HS72 detected in each of the 9 salt steps; in the other tubes, those peptides were only detected in limited salt steps (e.g., fraction/step/hits: 30/4/6, 29/2/2, 28/3/35, 27/8/ 55, 26/4/35, 24/8/119). Therefore, the detection and identification of other proteins, especially low-abundant proteins, became easier. IM23_YEAST (CAI ) 0.117), a low-abundance Journal of Proteome Research • Vol. 4, No. 4, 2005 1259

research articles

Li et al.

Figure 5. Correlation between the experimental and theoretical pI value of the proteins. Figure 4. Protein distribution in liquid IEF. The X-axis is the fraction number of RF3 fractionation. The Y-axis is the total peptide number of each fraction identified by 2D-LC/MS/MS. HS71_YEAST was concentrated in the acid fraction. ENO1_YEAST was concentrated in the neutral fraction. KPY1_YEAST was concentrated in the basic fraction.

protein, was identified based on one peptide (TPTDDANAAVGGQDTTKPK) in our approach of total identification (directly using 2D-LC-MS/MS). It was observed that the peptide was detected only in one SCX step. But, after the 2D-LC-MS/MS was combined with LIEF fractionation, this peptide was identified in a greater number of fractions of LIEF (19, 25, 26, 27, 29). Furthermore, other peptides from IM23_YEAST were also detected by MS, which improved the reliability of the identification of this protein. For example, in salt steps 3 and 4 of fraction 26, two unique peptides and five total hits of IM23_YEAST were detected (TPTDDANAAVGGQDTTKPK; LHPLAGLDK). In step 2 of fraction 27, 1 unique peptide detected, with 5 hits. In another example, MLC1_YEAST (P53141, myosin-2 light chain, CAI ) 0.217), one peptide (AIGYNPTNQLVQDIINADSSLR) was detected by MS, with 2 hits during total identification. However, after LIEF pre-fractionation was coupled with 2D-LC-MS/MS identification, the peptide was detected with 58 hits, indicating a more solid identification of the peptide. In salt steps 1, 2, 3 of fraction 25, 15 peptide hits from 4 unique peptides were detected (GVEVDSNGEIDYK, AIGYNPTNQLVQDIINADSSLR, DASSLTLDQITGLIEVNEK; DIFTLFDK). In steps 1, 2, 3, 4, 6, 7 of fraction 27 10 peptide hits were observed from 4 unique peptides (AIGYNPTNQLVQDIINADSSLR, GVEVDSNGEIDYK, LTDAEVDELLK, DASSLTLDQITGLIEVNEK). Many peptides were also detected in fractions 26 and 28 (YMLTGLGEK.L, YMLTGLGEKLTDAEVDELLK). Protein Distribution in LIEF. One major concern about liquid-phase IEF is that the proteins cannot isoelectrically focus as they should; this is due to diffusion of the proteins in solution. As is seen in Figure 4, most proteins with certain pI values were focused in the corresponding pH range of fractions and only few of them diffused into other fractions, indicating the focusing of proteins was acceptable. To evaluate the focusing by liquid IEF, we analyzed the correlation between the theoretical pI value and the experimental pI value. The theoretical pI values of corresponding proteins were obtained according to its primary structure form the website: http://www.expasy.ch. In liquid IEF fractionation, most of the proteins were found distributed in one or more fractions rather than only in a single fraction. The identified peptides in each fraction indicate the protein abundance in this fraction. It is believed that the protein should be distributed around the theoretical pI fraction corresponding to the protein abundance. Therefore, the experimental pI was calculated 1260

Journal of Proteome Research • Vol. 4, No. 4, 2005

Figure 6. Codon adaptation index (CAI) distribution of the identified S. cerevisiae proteome and the predicted S. cerevisiae genome. CAI distribution of the proteins predicted in the S. cerevisiae genome, identified by the method of LIEF-2D-LCMS/MS and by 2D-LC-MS/MS.

taking the protein distribution in account, as the formula listed below 30

∑(N × i

exppI

)

MpIi)

i)1

30

∑N

i

i)1

In this formula, exppI is the experimental pI value, N is the number of peptides of each protein in every fraction from 1 to 30, i is the fraction number of LIEF, and MpI is the mensurated pI value of the corresponding fraction measured by pH apparatus. We thus calculated the experimental pI value of every protein in our data. Then, according to the distribution of peptides from certain proteins in each fraction, we obtained the correlation between the experimental and theoretical pI value. Figure 5 shows the correlation of experimental pI and theoretical pI. The coefficient of high-abundance proteins was >0.8 if the identified peptide reached about 20 hits, indicating that high-abundance proteins were focused efficiently using this method. Protein Features Obtained by LIEF. Of the 6254 proteins in the yeast genome, 83.23% have CAI values between 0 and 0.20, which are predicted to be low-abundant (Figure 6). As shown in Figure 6, the data yields a representative sample of the yeast proteome with 1103 or 61.45% of the proteins identified having a CAI of 10 were identified, as well as the most basic protein, RL18_YEAST (P07279, pI 11.70553). In addition, proteins with MW < 10 000 and >180 000 Da are represented. For example, of the 6254 proteins in the yeast database, 9.35% (585 proteins) have MW values between 100 and 180 kDa, and 1.9% (119 proteins) have MW values in excess of 180 kDa (Figure 7. A). 13.04% (234 proteins) with MW values between 100 and 180 kDa, and 3.12% (56 proteins) with MW values in excess of 180 kDa were found (Figure 7. A). Figure 7B Shows the percentage of the proteins of every MW range in total proteins. For example, 1530 proteins have MW values between 30 and 50 kDa. In the yeast database (6254), the percent of this MW range is 24.46%. 456 proteins of our identified proteins (1795) have MW values between 30 and 50 kDa, the percent of this MW range is 25.91%. There is no obvious bias to this kind of protein in 2DE. In addition, we can find some proteins with MW < 10 000, 23 out of 168 potential proteins with MW < 10 kDa were identified, the smallest being R29A_YEAST (P41057) with a MW of 6525.25 Da.

without fractionation before 2D-LC-MS/MS. By the LIEF-2DLC-MS/MS combined method, we can improve the number of the low-abundance proteins to 5.43-fold, compared to only

We used 1%OG in the lysis buffer when preparing the sample. Using this procedure, both cytosolic and membrane proteins were available for analysis.9,29 Therefore, other extremes of the S. cerevisiae proteome are well represented in

Figure 8. Method of LIEF-2D-LC-MS/MS can found more proteins with extreme properties. From left to right are the percentages identified of total proteins, proteins with a CAI < 0.2, proteins with a pI < 4.3, proteins with a pI > 11, proteins with a MW < 10kDa, proteins with a MW > 180 kDa, proteins with a GRAVY > 0, transmember (TM) with one or more predicted transmembrane domains, transmember (TM) with three or more predicted transmembrane domains. Journal of Proteome Research • Vol. 4, No. 4, 2005 1261

research articles

Li et al.

the classification of identified proteins with CAI under 0.2. Enzyme identifications accounted for the largest percentage (Figure 9 A. 33.1%, Figure 9 B. 31.54%). Approximately 20.77% and 24.72% of unknown or hypothetical proteins were identified in Figure 9A,B. The percentage of signal transducers and unknown proteins in certain total proteins were increased. The results reported in this paper were from 30 fractions, collected simultaneously in one LIEF run. After combining the MS/MS data generated from all three samples, we were able to identify 1795 proteins from the S. cerevisiae proteome. A complete list of the proteins and peptides identified is available in Supporting Table 1. This analysis revealed that the LIEF is a preparative scale method, which can load as much as 20 mg. The proteins are separated on the basis of the pI distribution of proteins as well as charge and hydrophobicity. Our LIEF-2D-LC-MS/MS method is largely unbiased, meaning that low-abundance proteins, proteins with extreme pI and MW, and integral membrane proteins were identified.

Conclusion

Figure 9. Classification of identified proteins obtained by LIEF2D-LC-MS/MS. Classification of all identified proteins is showed in (A). Classification of identified proteins with CAI under 0.2 is showed in (B).

our data, compared to 2-DE, which cannot apply good resolution for hydrophobic proteins. More hydrophobic proteins have been found in our study in the liquid mode. Proteins with both hydrophobic and hydrophilic GRAVYs are also represented in our data set. 125 proteins with GRAVY > 0 was identified, the largest one being VATL_YEAST (P25515) with a GRAVY of 1.202. Although the 1795 proteins we identified do not represent a complete analysis of all the proteins present in logarithmically growing cells, our method clearly provides a large-scale and global view of the S. cerevisiae proteome. Our methodology not only gives access to low-abundance proteins, membrane proteins, proteins with MW in excess of 180 kDa, proteins with GRAVY > 0, and proteins with pIs > 10, but more importantly, it performed in an unbiased manner. Figure 9 illustrates this point by plotting the number of proteins identified in a particular class, as a percentage of the predicted proteins. The sensitivity level across the classes of proteins listed ranged from 11.7% of the predicted proteins identified with GRAVYs > 0 to 47.06% of the predicted proteins identified with MW > 180 kDa (Figure 8). Furthermore, we identified 121 proteins with three or more predicted transmembrane domain. The method has a slight bias against proteins with a pI < 4.3 and MW > 180 kDa, although proteins from both of theses classes were identified. Functional Classification of the Total Proteins. A total of 1795 different proteins were identified within the yeast proteome. All identified proteins in this study were classified into several groups including signal transducers, structural molecules, transporters, enzymes, enzyme regulators, ligand bindings or carriers, unknown proteins, transcription regulators, based on their functional categories in the “SWISS-PROT and TrEMBL” protein database (Figure 9). Figure 9 A shows the classification of all identified proteins and Figure 9. B shows 1262

Journal of Proteome Research • Vol. 4, No. 4, 2005

LIEF-2D-LC-MS/MS has been shown to be a feasible method for the preparative separation of large numbers of proteins from yeast total lysate. The method is capable of fractionate protein according to their pI in a liquid phase, and the resulted protein fractions can be directly digested and identified by 2D-LC-MS/MS. There were 1795 unique proteins identified by 2D-LC-MS/MS, containing a number of lowabundance proteins. A main advantage of this liquid 3-D separation is that the liquid IEF is preparative scale, which can be loaded with as much as 20 mg of protein. Also, This method can be widely used on similar instrument with LIEF function and is potential to fractionate more complex proteome such as plasma with wide dynamic range. Abbreviations: RF3, Recycling free-flow focusing; 2DE, twodimensional electrophoresis; LIEF, liquid isoelectric focusing; MCE, multi-compartment electrolyzer; MudPIT, multidimensional protein identification technology; GRAVY, Grand average of hydropathicity; TM, transmembrane domain; CAI, code adaptation index.

Acknowledgment. This work was supported by the National High-Technology Project (2002BA711A11) and Basic Research Foundation (2001CB210501, 2002CB713807). References (1) Issaq, H. J., The role of separation science in proteomics research. Electrophoresis 2001, 22 (17), 3629-3638. (2) Issaq, H. J.; Conrads, T. P.; Janini, G. M.; Veenstra, T. D., Methods for fractionation, separation and profiling of proteins and peptides. Electrophoresis 2002, 23 (17), 3048-3061. (3) Righetti, P. G.; Castagna, A.; Herbert, B.; Reymond, F.; Rossier, J. S., Prefractionation techniques in proteome analysis. Proteomics 2003, 3 (8), 1397-1407. (4) O’Farrell, P. H., High-resolution two-dimensional electrophoresis of proteins. Midwifery 1975, 250 (10), 4007-4021. (5) Gorg, A.; Obermaier, C.; Boguth, G.; Harder, A.; Scheibe, B.; Wildgruber, R.; Weiss, W., The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 2000, 21 (6), 1037-1053. (6) Neidhardt, F. C.; Appleby, D. B.; Sankar, P.; Hutton, M. E.; Phillips, T. A., Genomically linked cellular protein databases derived from two-dimensional polyacrylamide gel electrophoresis. Electrophoresis 1989, 10 (2), 116-122. (7) Anderson, N. L.; Esquer-Blasco, R.; Hofmann, J. P.; Anderson, N. G., A two-dimensional gel database of rat liver proteins useful in gene regulation and drug effects studies. Electrophoresis 1991, 12 (11), 907-930.

Prefractionation of Proteome by LIEF and 2D-LC-MS/MS (8) Patterson, S. D., Matrix-assisted laser-desorption/ionization mass spectrometric approaches for the identification of gel-separated proteins in the 5-50 pmol range. Electrophoresis 1995, 16 (7), 1104-1114. (9) Wall, D. B.; Kachman, M. T.; Gong, S.; Hinderer, R.; Parus, S.; Misek, D. E.; Hanash, S. M.; Lubman, D. M., Isoelectric focusing nonporous RP HPLC: a two-dimensional liquid-phase separation method for mapping of cellular proteins with identification using MALDI-TOF mass spectrometry. Anal. Chem. 2000, 72 (6), 10991111. (10) Shevchenko, A.; Wilm, M.; Vorm, O.; Mann, M., Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal. Chem. 1996, 68 (5), 850-858. (11) Swain, M.; Ross, N. W., A silver stain protocol for proteins yielding high resolution and transparent background in sodium dodecyl sulfate-polyacrylamide gels. Electrophoresis 1995, 16 (6), 948951. (12) McDonald, W. H.; Yates, J. R., 3rd, Shotgun proteomics and biomarker discovery. Dis. Markers 2002, 18 (2), 99-105. (13) Wu, C. C.; MacCoss, M. J., Shotgun proteomics: tools for the analysis of complex biological systems. Curr. Opin. Mol. Ther. 2002, 4 (3), 242-250. (14) Hancock, W. S.; Wu, S. L.; Shieh, P., The challenges of developing a sound proteomics strategy. Proteomics 2002, 2 (4), 352-359. (15) Phizicky, E.; Bastiaens, P. I.; Zhu, H.; Snyder, M.; Fields, S., Protein analysis on a proteomic scale. Nature 2003, 422 (6928), 208-215. (16) Pandey, A.; Mann, M., Proteomics to study genes and genomes. Nature 2000, 405 (6788), 837-846. (17) Link, A. J.; Eng, J.; Schieltz, D. M.; Carmack, E.; Mize, G. J.; Morris, D. R.; Garvik, B. M.; Yates, J. R., 3rd, Direct analysis of protein complexes using mass spectrometry. Nat Biotechnol 1999, 17 (7), 676-682. (18) Washburn, M. P.; Wolters, D.; Yates, J. R., 3rd, Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 2001, 19 (3), 242-247. (19) Eng JK, M. A., Yates JR III, 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, 976-989. (20) Gygi, S. P.; Rist, B.; Gerber, S. A.; Turecek, F.; Gelb, M. H.; Aebersold, R., Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nat. Biotechnol. 1999, 17 (10), 994-999. (21) Kislinger, T.; Emili, A., Going global: protein expression profiling using shotgun mass spectrometry. Curr. Opin. Mol. Ther. 2003, 5 (3), 285-293. (22) Yates, J. R., 3rd, Mass spectrometry. From genomics to proteomics. Trends Genet. 2000, 16 (1), 5-8. (23) Nilsson, C. L.; Puchades, M.; Westman, A.; Blennow, K.; Davidsson, P., Identification of proteins in a human pleural exudate using two-dimensional preparative liquid-phase electrophoresis and matrix-assisted laser desorption/ionization mass spectrometry. Electrophoresis 1999, 20 (4-5), 860-865. (24) Hesse, C.; Nilsson, C. L.; Blennow, K.; Davidsson, P., Identification of the apolipoprotein E4 isoform in cerebrospinal fluid with preparative two-dimensional electrophoresis and matrix assisted laser desorption/ionization-time-of-flight-mass spectrometry. Electrophoresis 2001, 22 (9), 1834-1837. (25) Ho, Y.; Gruhler, A.; Heilbut, A.; Bader, G. D.; Moore, L.; Adams, S. L.; Millar, A.; Taylor, P.; Bennett, K.; Boutilier, K.; Yang, L.; Wolting, C.; Donaldson, I.; Schandorff, S.; Shewnarane, J.; Vo, M.; Taggart, J.; Goudreault, M.; Muskat, B.; Alfarano, C.; Dewar, D.; Lin, Z.; Michalickova, K.; Willems, A. R.; Sassi, H.; Nielsen, P. A.; Rasmussen, K. J.; Andersen, J. R.; Johansen, L. E.; Hansen, L. H.; Jespersen, H.; Podtelejnikov, A.; Nielsen, E.; Crawford, J.; Poulsen, V.; Sorensen, B. D.; Matthiesen, J.; Hendrickson, R. C.; Gleeson, F.; Pawson, T.; Moran, M. F.; Durocher, D.; Mann, M.; Hogue, C. W.; Figeys, D.; Tyers, M., Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature 2002, 415 (6868), 180-183. (26) Shang, T. Q.; Ginter, J. M.; Johnston, M. V.; Larsen, B. S.; McEwen, C. N., Carrier ampholyte-free solution isoelectric focusing as a prefractionation method for the proteomic analysis of complex protein mixtures. Electrophoresis 2003, 24 (14), 2359-2368. (27) Bier, M., Recycling isoelectric focusing and isotachophoresis. Electrophoresis 1998, 19 (7), 1057-1063. (28) Kachman, M. T.; Wang, H.; Schwartz, D. R.; Cho, K. R.; Lubman, D. M., A 2-D liquid separations/mass mapping method for interlysate comparison of ovarian cancers. Anal. Chem. 2002, 74 (8), 1779-1791.

research articles (29) Wall, D. B.; Kachman, M. T.; Gong, S. S.; Parus, S. J.; Long, M. W.; Lubman, D. M., Isoelectric focusing nonporous silica reversedphase high-performance liquid chromatography/electrospray ionization time-of-flight mass spectrometry: a three-dimensional liquid-phase protein separation method as applied to the human erythroleukemia cell-line. Rapid Commun. Mass Spectrom. 2001, 15 (18), 1649-1661. (30) Wall, D. B.; Parus, S. J.; Lubman, D. M., Comparison of the capabilities of liquid isoelectric focusing-one-dimensional nonporous silica reversed-phase liquid chromatography-electrospray ionization time-of-flight mass spectrometry and liquid isoelectric focusing-one-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis mass mapping for the analysis of intact protein molecular masses. J. Chromatogr. B Biomed. Sci. Appl. 2001, 763 (1-2), 139-148. (31) Wang, H.; Kachman, M. T.; Schwartz, D. R.; Cho, K. R.; Lubman, D. M., A protein molecular weight map of ES2 clear cell ovarian carcinoma cells using a two-dimensional liquid separations/mass mapping technique. Electrophoresis 2002, 23 (18), 3168-3181. (32) Wang, M. Z.; Howard, B.; Campa, M. J.; Patz, E. F., Jr.; Fitzgerald, M. C., Analysis of human serum proteins by liquid-phase isoelectric focusing and matrix-assisted laser desorption/ionizationmass spectrometry. Proteomics 2003, 3 (9), 1661-1666. (33) Righetti, P.; Wenisch, E.; Faupel, M., Preparative protein purification in a multi-compartment electrolyser with immobiline membranes. J. Chromatogr. A 1989, 475 (2), 293-309. (34) Righetti, P. G.; Wenisch, E.; Jungbauer, A.; Katinger, H.; Faupel, M., Preparative purification of human monoclonal antibody isoforms in a multi-compartment electrolyser with immobiline membranes. J. Chromatogr. 1990, 500, 681-696. (35) Herbert, B.; Righetti, P. G., A turning point in proteome analysis: sample prefractionation via multicompartment electrolyzers with isoelectric membranes. Electrophoresis 2000, 21 (17), 3639-3648. (36) Zuo, X.; Speicher, D. W., A method for global analysis of complex proteomes using sample prefractionation by solution isoelectrofocusing prior to two-dimensional electrophoresis. Anal. Biochem. 2000, 284 (2), 266-278. (37) Zuo, X.; Speicher, D. W., Comprehensive analysis of complex proteomes using microscale solution isoelectrofocusing prior to narrow pH range two-dimensional electrophoresis. Proteomics 2002, 2 (1), 58-68. (38) Egen, N. B.; Cuevas, W. A.; McNamara, P. J.; Sammons, D. W.; Humphreys, R.; Songer, J. G., Purification of the phospholipase D of Corynebacterium pseudotuberculosis by recycling isoelectric focusing. Am. J. Vet. Res. 1989, 50 (8), 1319-1322. (39) Svensson, H., Acta Chem. Scand 1961, 15, 325-341. (40) Svensson, H., Acta Chem. Scand 1961, 16, 456-466. (41) Carr, S.; Aebersold, R. Mol. Cell Proteomics 2004, 3 (6), 531-533. (42) Nagabhushan., T. L., Electrophoresis 1986, 7, 552-557. (43) Zeng, R.; Chen, Y. B.; Shao, X. X.; Shieh, C. H.; Miller, K.; Tran, H.; Xia, Q. C., Identification of proteins separated by onedimensional sodium dodecyl sulfate/polyacrylamide gel electrophoresis with matrix-assisted laser desorption/ionization ion trap mass spectrometry; comparison with matrix-assisted laser desorption/ionization time-of-flight mass fingerprinting. Rapid Commun. Mass Spectrom. 2003, 17 (17), 1995-2004. (44) Yu, L. R.; Zeng, R.; Shao, X. X.; Wang, N.; Xu, Y. H.; Xia, Q. C., Identification of differentially expressed proteins between human hepatoma and normal liver cell lines by two-dimensional electrophoresis and liquid chromatography-ion trap mass spectrometry. Electrophoresis 2000, 21 (14), 3058-3068. (45) Yu, L. R.; Shao, X. X.; Jiang, W. L.; Xu, D.; Chang, Y. C.; Xu, Y. H.; Xia, Q. C., Proteome alterations in human hepatoma cells transfected with antisense epidermal growth factor receptor sequence. Electrophoresis 2001, 22 (14), 3001-3008. (46) Langen, H.; Roder, D.; Juranville, J. F.; Fountoulakis, M., Effect of protein application mode and acrylamide concentration on the resolution of protein spots separated by two-dimensional gel electrophoresis. Electrophoresis 1997, 18 (11), 2085-2090. (47) Ma, J.; Stoter, G.; Verweij, J.; Schellens, J. H., Comparison of ethanol plasma-protein precipitation with plasma ultrafiltration and trichloroacetic acid protein precipitation for the measurement of unbound platinum concentrations. Cancer Chemother. Pharmacol. 1996, 38 (4), 391-394.

Journal of Proteome Research • Vol. 4, No. 4, 2005 1263

research articles (48) Wade, C. R.; van Rij, A. M., Plasma thiobarbituric acid reactivity: reaction conditions and the role of iron, antioxidants and lipid peroxy radicals on the quantitation of plasma lipid peroxides. Life Sci. 1988, 43 (13), 1085-1093. (49) Chemale, G.; van Rossum, A. J.; Jefferies, J. R.; Barrett, J.; Brophy, P. M.; Ferreira, H. B.; Zaha, A., Proteomic analysis of the larval stage of the parasite Echinococcus granulosus: causative agent of cystic hydatid disease. Proteomics 2003, 3 (8), 1633-1636. (50) Nandakumar, M. P.; Shen, J.; Raman, B.; Marten, M. R., Solubilization of trichloroacetic acid (TCA) precipitated microbial proteins via naOH for two-dimensional electrophoresis. J. Proteome Res. 2003, 2 (1), 89-93. (51) Polson, C.; Sarkar, P.; Incledon, B.; Raguvaran, V.; Grant, R., Optimization of protein precipitation based upon effectiveness of protein removal and ionization effect in liquid chromatography-tandem mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2003, 785 (2), 263-275. (52) Park, Z. Y.; Russell, D. H., Identification of individual proteins in complex protein mixtures by high-resolution, high-mass-accuracy MALDI TOF-mass spectrometry analysis of in-solution thermal denaturation/enzymatic digestion. Anal. Chem. 2001, 73 (11), 2558-2564.

1264

Journal of Proteome Research • Vol. 4, No. 4, 2005

Li et al. (53) Lee, C. L.; Hsiao, H. H.; Lin, C. W.; Wu, S. P.; Huang, S. Y.; Wu, C. Y.; Wang, A. H.; Khoo, K. H., Strategic shotgun proteomics approach for efficient construction of an expression map of targeted protein families in hepatoma cell lines. Proteomics 2003, 3 (12), 2472-2486. (54) McDonald, W. H.; Yates, J. R., 3rd, Shotgun proteomics: integrating technologies to answer biological questions. Curr. Opin. Mol. Ther. 2003, 5 (3), 302-309. (55) Xiao, Z.; Conrads, T. P.; Lucas, D. A.; Janini, G. M.; Schaefer, C. F.; Buetow, K. H.; Issaq, H. J.; Veenstra, T. D., Direct ampholytefree liquid-phase isoelectric peptide focusing: application to the human serum proteome. Electrophoresis 2004, 25 (1), 128-133. (56) Janini, G. M.; Conrads, T. P.; Veenstra, T. D.; Issaq, H. J., Development of a two-dimensional protein-peptide separation protocol for comprehensive proteome measurements. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2003, 787 (1), 4351. (57) Futcher, B.; Latter, G. I.; Monardo, P.; McLaughlin, C. S.; Garrels, J. I., A sampling of the yeast proteome. Mol Cell Biol 1999, 19 (11), 7357-7368.

PR049751G