Consecutive Proteolytic Digestion in an Enzyme Reactor Increases

Article pubs.acs.org/ac

Consecutive Proteolytic Digestion in an Enzyme Reactor Increases Depth of Proteomic and Phosphoproteomic Analysis Jacek R. Wiśniewski* and Matthias Mann* Department of Proteomics and Signal Transduction, Max Planck Institute of Biochemistry, D-82152 Martinsried, Germany S Supporting Information *

ABSTRACT: Analytical advantages of using multiple enzymes for sample digestion (MED), primarily an increase of sequence coverage, have been reported in several studies. However, this approach is only rarely used, mainly because it requires additional sample and mass spectrometric measurement time. We have previously described Filter Aided Sample Preparation (FASP), a type of proteomic reactor, in which samples dissolved in sodium dodecyl sulfate (SDS) are digested in an ultrafiltration unit. In FASP, such as in any other preparation protocol, a portion of sample remains after digestion and peptide elution. Making use of this fact, we here develop a protocol enabling consecutive digestion of the sample with two or three enzymes. By use of the FASP method, peptides are liberated after each digestion step and remaining material is subsequently cleaved with the next proteinase. We observed excellent performance of the ultrafiltration devices in this mode, allowing efficient separation of orthogonal populations of peptides, resulting in an increase in the numbers of identified peptides and proteins. At the low microgram level, we found that the consecutive use of endoproteinases LysC and trypsin enabled identification of up to 40% more proteins and phosphorylation sites in comparison to the commonly used one-step tryptic digestion. MED-FASP offers efficient exploration of previously unused sample material, increasing depth of proteomic analyses and sequence coverage.

I

material and without apparent biases against any protein classes.4 FASP typically converts more than 50% of the protein mass to peptides.4,13,14 Here, we describe an approach allowing digestion of the reaming protein material with a second or third enzyme that is sequentially applied to the reactor. The FASP method allows stepwise elution of peptides liberated after each digestion step. This multienzyme digestion (MED) FASP approach appears to be especially useful for the analysis of samples available only in minute amounts, significantly increasing the number of identified proteins and their sequence coverage. We also demonstrate increased depth of identification of phosphorylation sites.

n bottom-up proteomic experiments, the depth of proteome characterization and the sequence coverage of individual proteins improve with the number of different peptides that are mass-measured and fragmented. Although identification of peptides primarily depends on the sensitivity and speed of regular and tandem mass spectrometric (MS and MS/MS) measurements, which is a feature of the mass spectrometer used, the nature of the peptide mixture to be analyzed also plays a pivotal role. In the vast majority of proteomic studies, peptides are prepared by digestion of proteins with a single enzyme. Most frequently trypsin and to a lesser extent endoproteinases LysC are employed. Sometimes, the tryptic digest is preceded by LysC digestion, which can be performed under harsher conditions, but this procedure still results in a single peptide population. Several studies have shown that parallel digestion and analysis of aliquots of the same sample with different enzymes can increase the range of peptides created. As a result, protein sequence coverage as well as the number of identified proteins improves.1−6 On the other hand, complementary digestion has obvious limitations that include the requirement of higher sample amounts and increased measuring time. Digestion of proteomes to peptides is usually done in gel7 or in solution;8,9 however, it can also be performed in a “proteomic reactor” format.10−12 We have described the filter-aided sample preparation (FASP), in which samples are completely dissolved in sodium dodecyl sulfate (SDS), which is exchanged to urea on top of a filter device, and which consequently leads to the generation of clean peptides from a wide variety of input © 2012 American Chemical Society

■

EXPERIMENTAL SECTION HeLa Cells and Tissue Lysates. Cell pellets were flashfrozen in liquid nitrogen and stored at −80 °C. Cells were lysed in a buffer consisting of 0.1 M tris(hydroxymethyl)aminomethane hydrochloride (Tris-HCl), pH 8.0, 0.1 M dithiothreitol (DTT), and 2% SDS at 99 °C for 5 min. Brain and liver from C57BL/6 mouse were lysed in 2% SDS and 0.1 M DTT in 0.1 M Tris-HCl, pH 7.6, as described previously.13 After being chilled to room temperature, the lysates were sonicated with a Branson-rod-type and then were clarified by centrifugation at 16100g for 10 min. Because of the presence of Received: September 23, 2011 Accepted: February 12, 2012 Published: February 12, 2012 2631

dx.doi.org/10.1021/ac300006b | Anal. Chem. 2012, 84, 2631−2637

Analytical Chemistry

Article

Table 1. Digestion Conditions Used supplier

protein to enzyme ratio

trypsin LysC ArgC

Promega Wako Roche

1:100 1:50 1:50

GluC AspN chymotrypsin

Roche Roche Roche

1:50 1:50 1:50

enzyme

buffer

digestion time, h

digestion temp, °C

40 mM (NH)4HCO3 50 mM Tris-HCl, pH 8.5 50 mM Tris-HCl, pH 8.5, 10% activation solution (Roche), and 5 mM CaCl2 40 mM (NH)4HCO3 40 mM (NH)4HCO3 50 mM Tris-HCl, pH 7.8, and 10 mM CaCl2

18 18 18

37 30 37

18 18 4

20 37 30

Identification of Phosphorylation Sites. Aliquots of brain and liver lysates containing 1.6 and 2.2 mg of total protein were processed in 0.5 mL Amicon Ultra 30k filtration units according to the FASP procedure by use of the two-step protocol described above. Pure acetonitrile and 10% (v/v) CF3COOH were added to the digests to final concentrations of 30% (v/v) and 1% (v/v), respectively. Phosphopeptides were enriched on TiO2 beads according to ref 17 with minor modifications as described previously.13 The ratio of TiO2 beads to the total peptides was 2:1. LC-MS/MS Analysis. Liquid chromatography was performed on a Proxeon Easy-nLC System (Proxeon Biosystems, Odense, Denmark; now Thermo Fisher Scientific). Peptides were separated on a 15 cm fused silica emitter (Proxeon Biosystems) packed in-house with the reverse-phase material ReproSil-Pur C18-AQ, 3 μm resin (Dr. Maisch GmbH, Ammerbuch-Entringen, Germany) with a 230 min gradient from 2% to 80% of 80% (v/v) CH3CN and 0.5% (v/v) acetic acid. The MS/MS analysis of HeLa cells was carried out on an LTQ-Orbitrap instrument (Thermo Fisher Scientific), whereas phosphorylation site mapping was performed on an LTQ-Orbitrap Velos mass spectrometer. Survey MS scans were acquired in the Orbitrap analyzer with a resolution of 60 000 at m/z 400. In the LTQOrbitrap XL full mass range scans at a resolution of 60 000 were acquired while up to 10 MS/MS spectra were acquired at low resolution in the linear ion trap (“top10” method). On the LTQOrbitrap Velos, MS scans were acquired in a high−high strategy with higher energy collisional dissociation (HCD).18 For HCD analysis, transients of 96 ms were acquired, corresponding to a resolution of 7500 at m/z 400. Data Analysis. The MS data were analyzed in the software environment MaxQuant19 version 1.1.1.27 using the Andromeda search engine.20 Proteins were identified by searching MS and MS/MS data of peptides against a decoy version of the International Protein Index (IPI) database for human (v. 3.68). Data obtained by analysis of peptide mixtures digested with trypsin after the first digestion with GluC or AspN were searched for mixed specificity, GluC/trypsin or AspN/trypsin, respectively. For peptides obtained with chymotrypsin following tryptic digestion, mixed trypsin/chymotrypsin specificity was used. Carbamidomethylation of cysteines was set as a fixed modification. The minimum peptide length was specified to be 6 amino acids. The initial maximal mass tolerance in MS mode was set to 7 ppm, whereas fragment mass tolerance was set to 0.5 Th for CID data and 20 ppm for HCD data. The maximum false positive rate (FDR) for peptides and for proteins was specified as 0.01.

high levels of SDS in the FASP protocol, all lysates were prepared in the absence of protease or phosphatase inhibitors. Total Protein and Peptide Determination. Protein content was determined on a Cary Eclipse fluorescence spectrometer (Varian, Palo Alto, CA) as described previously.15 Tissue lysates were assayed by adding 1−2 μL of sample or tryptophan standard mixed with 2 mL of 8 M urea in 10 mM Tris-HCl, pH 8.0. The peptides resulting from FASP digests were analyzed in 0.2 mL of buffer used for peptide elution in 5 × 5 mm quartz cells. Fluorescence was measured at 295 nm for excitation and 350 nm for emission. The slits were set to 5 and 20 nm for excitation and emission, respectively. Protein Multidigestion by FASP. Detergent was removed from the lysates and the proteins were digested with trypsin via the FASP protocol4 with spin ultrafiltration units of nominal molecular weight cutoff 30 000. Briefly, to YM-30 microcon filter units (catalogue no. MRCF0R030, Millipore) containing protein concentrates, 200 μL of 8 M urea in 0.1 M Tris-HCl, pH 8.5 (UA buffer) was added, and samples were centrifuged at 14000g at 20 °C for 15 min. This step was repeated once. Then 50 μL of 0.05 M iodoacetamide in 8 M urea was added to the filters and the samples were incubated in darkness for 10 min. Filters were washed twice with 100 μL of UA buffer followed by two washes with 100 μL of 40 mM NH4HCO3. Protein digestions were conducted overnight with endoproteinases GluC, ArgC, LysC, AspN, trypsin, and chymotrypsin. After digestion, the liberated peptides were collected by centrifugation and the filtration units were washed once with 40 μL of UA buffer and subsequently with two 40 μL washes with water, before the next digestion step. Conditions of the digestions are summarized in Table 1. Peptide concentrations were determined as described above. Peptide Fractionation. Peptides were fractionated according to the previously described pipet-tip strong anion exchange (SAX) protocol16 with minor modifications. Briefly, peptides were loaded into tip columns made by stacking six layers of a 3 M Empore anion exchange disk (1214−5012, Varian, Palo Alto, CA) into a 200 μL micropipet tip. For column equilibration and elution of fractions, we used Britton & Robinson universal buffer (BRUB) composed of 20 mM acetic acid, 20 mM phosphoric acid, and 20 mM boric acid titrated with NaOH to the desired pH. In the experiments where LysC was used, peptides were loaded into the pipet-tip column at pH 11 and the three fractions were eluted at pH 6, 4, and 2. Tryptic peptides were loaded at pH 5 and eluted at pH 2. The flow-through and the fractions were analyzed. The peptides resulting from chymotryptic digestion were analyzed directly by liquid chromatrography (LC)-MS/MS. The peptides obtained by digestion only with trypsin were loaded at pH 11, and fractions were subsequently eluted with buffer solutions of pH 8, 6, 5, 4, and 2 in turn.

■

RESULTS AND DISCUSSION Consecutive Enzyme Digestion in FASP Format. Filteraided sample preparation allows efficient digestion of proteins

2632



Article

with different enzymes and separation of the peptides from longer polypeptides or undigested nonproteinaceous material by ultrafiltration.4 The material retained on the filter after digestion with a first enzyme is normally discarded, but we reasoned that it can be subjected to further digestion steps with a second enzyme with different cleavage specificity, which should liberate more of the material (Figure 1). This procedure can be repeated

Figure 2. Sequential digestion of samples with two enzymes by FASP. HeLa lysates containing 25 μg of total protein were digested with endoproteinases GluC, LysC, ArgC, AspN, and trypsin. Following elution of the released peptides, the material remaining in the ultrafiltration devices was digested with trypsin and the obtained peptides were collected. The experiments were performed in triplicate. The peptide yields (A) were determined by fluorometry as described in the Experimental Section, and the fractions were analyzed by LCMS/MS in separate runs. The numbers of identified peptides and proteins in individual fractions are shown in panels B and C. Error bars show standard deviations.

Figure 1. Principle of multiple-step enzyme digestion filter-aided sample preparation (MED FASP). (A) Detergent-containing samples are processed according to the FASP procedure, allowing complete detergent depletion. In the first step proteins are digested with an enzyme A and the released peptides are eluted. Then the remaining high molecular weight material is digested by an enzyme B that has distinct cleavage specificity from enzyme A. Following elution of peptides (peptides B), the polypeptide residual can be digested with an enzyme C which is different than A and B. The obtained peptides are analyzed directly by separate LC-MS/MS runs from samples containing up to several micrograms of protein or for larger samples are prefractionated by SAX before LC-MS/MS. (B) Illustration of the advantages of sequential digestion and peptide elution (MED FASP) over parallel digestion of a sample with the two enzymes LysC and trypsin. Parallel digestion results in the same mixtures of digestion product from two enzymes. In contrast, a successive digestion results in fractions containing different peptides.

cleavage only. Each sample contained 25 μg of total protein. Digestion with GluC, ArgC, LysC, and AspN yielded on average 6.2, 6.4, 8.8, and 6.1 μg of total peptide, and the postdigestion with trypsin yielded 5.8, 5.2, 3.7, and 6.6 μg of total peptide, respectively (Figure 2A). In the control digestion with trypsin, 13.8 μg of peptides was obtained in the first digestion and 1.8 μg in the second digestion (Figure 2A). The individual peptide mixtures were analyzed by LC-MS/ MS and the data were searched by MaxQuant. The results obtained from the data analysis are shown in Supporting Information Tables 1 and 2 and are summarized in Figure 2B,C. The peptides generated by GluC, ArgC, and AspN provided the lowest number of peptide and protein identifications. The peptide numbers obtained from the LysC digests were 2 times higher but similar to those found after analysis of tryptic peptides. Next, we combined the data obtained in the analysis of the first and the second digestion. The digestion by LysC followed by trypsin led to the highest number of unique peptide identifications, whereas GluC or AspN followed by

for a third enzyme. In this way different populations of peptides can be obtained from a single sample at no cost in additional input material. To test the potential advantages of this strategy, we digested samples with three different combinations of two endoproteinases. For the first digestion, endoproteinases GluC, ArgC, AspN, and LysC were selected. The second digestion was carried out with trypsin, which is the most efficient and universal protease in proteomics (Figure 2). Peptides obtained in the first and second digestions were analyzed in separate LC-MS/MS runs. For comparison, two samples were subjected to tryptic 2633



Article

peptides obtained in the first digestion more frequently have small uncharged residues such as S, T, and M at position +1 with respect to the K/R than the peptides from the second digestion. In contrast, the latter have the positively charged histidine in position +1 about twice as often, when compared to the peptides from the first digestion. As a consequence of the generation of distinct populations of peptides, in two consecutive tryptic digestions a higher number of peptides and proteins can be identified, compared to a cumulative analysis of two parallel digestions, which additionally consumes 2 times more sample (Figure 3B). In all of the tested two-step digestions the total peptide yield was similar, which implies that the trypsin digestion compensated for less efficient primary digestion. Furthermore, the overlap of peptide populations was low in all combinations. However, the numbers of identified peptides (and as a consequence the number of proteins) were different. In particular, in the experiment with GluC and AspN the numbers were 2 times lower. This at least partly reflects lower MS/MS identification rates for peptides obtained by GluC and AspN digestions as well as those generated by trypsin in the second cleavage step. Whereas pure tryptic peptides were identified at a ratio of 50−60% of the MS/MS events, only 35−38% of the targeted GluC and AspN peptides resulted in successful identification (not shown). Similarly low identification rates were observed for tryptic peptides obtained from GluC and ApsN predigested sample, which is possibly due to the fact that the search space for peptides with tryptic and GluC or ApsN specificity is much larger. Finally, we compared the sequence coverage obtained by the two-step digestion methods (Supporting Information Table 3 and Figure 3B). The highest average sequence coverage per protein of 19.7% was obtained in the ArgC/trypsin digestion method. The combination of LysC and trypsin resulted in sequence coverage of 18.4%, whereas repeated digestion with trypsin covered on average 16.3% of the protein sequences. AspN/trypsin and GluC/ trypsin digestions resulted in peptide identifications covering only 12.2% and 12.4% of the identified proteins, respectively. In all combinations of digesting enzymes, the second digestion led to an increase of sequence coverage. When both the number of identified proteins and their sequence coverage were considered, the LysC/trypsin combination gave best results, and therefore it was generally applied in the subsequent experiments. MED FASP for Different Sample Amounts and for the Analysis of Very Small Cell Populations. Next we tested the applicability of two-stage digestion with LysC and trypsin for analysis of samples ranging in size from 2.8 to 37 μg of total protein (Figure 4 and Supporting Information Table 4). These protein amounts correspond to a range of about 2 × 104 to 2.5 × 105 HeLa cells. Over this sample range, the consecutive digest with LysC and trypsin resulted in a 50% yield of peptides from total protein input (Figure 4 A). The inset in Figure 4A shows that the amount of peptide was directly proportional to the amount of processed lysate, proving the applicability of the MED-FASP approach for analysis of minute amounts of samples. Analysis of the 15 and 37 μg samples enabled identification of, on average, 17 000 peptides and 3300 proteins (Figure 4B,C). The numbers obtained for 4.5 and 2.8 μg were lower but these sample amounts still allowed, on average, identification of 2900 and 2700 proteins, respectively (Figure 4C). MED FASP Combined with SAX Fractionation for Economical Sample Exploration. While many biological samples can be collected for proteomic experiments in large amounts, often economic issues, such as expensive media or

trypsin was less efficient (Figure 2B). On the level of identified proteins, the combination of LysC and trypsin led to the best result, with a total of 3350 ± 67 identified proteins. The control analysis of the tryptic peptides enabled identification of 2560 ± 29 proteins in combined searches of the two runs (Figure 2C). These results clearly show that consecutive digestion of the sample with LysC and trypsin results in a large increase in identified proteins. This can be attributed to excellent separation of the peptides obtained in the first digestion step from the material remaining on the ultrafiltration membrane, leading to nearly completely orthogonal peptide populations in the two analyses. In particular, only 0.05%, 1.1%, and 3.6% common peptides between the first and second digestion were observed in the AspN/trypsin, GluC/trypsin, and LysC/trypsin experiments, respectively (Figure 3A). Such nearly complete

Figure 3. (A) Venn plots of the data from Figure 2B show the overlap of peptides identified in the first digest with GluC, LysC, ArgC, AspN, or trypsin and the following second tryptic digest. “Parallel” digestion shows the overlap of two tryptic digests obtained in the first cleavage step. (B) Protein sequence coverage obtained with different enzymes in the digestion and its increase by tryptic postdigestion of the material remaining on the filter.

separation of two peptide populations is unusual and is difficult to achieve even with chromatographic methods. Two-step digestions with trypsin showed a 24.5% overlap between the peptides identified in the first digest and the postdigestion of the material remaining on the filter. The peptides obtained in these steps contain similar numbers of missed cleavages of, on average, 24% and 22% per identified peptide, and therefore the differences in both peptide populations probably reflect differences in the affinity of trypsin to cleave at K/R within various amino acid sequences. For example, 2634



Article

Figure 4. Multistep enzyme digestion can be applied to samples containing low and high protein amounts. Four different sample amounts of HeLa lysate were processed with FASP via the two-enzyme digestion protocol. Each sample was analyzed in duplicate. The yields of the peptides were determined by fluorometry (A), and the fractions were analyzed by LC-MS/MS (B and C). (*) Only 50% of the digests was used for LC-MS/MS analysis. Data points for sample containing 25 μg of total protein are from Figure 2.

Figure 5. Analysis of the HeLa proteome by trienzyme MED-FASP combined with SAX fractionation. HeLa lysates containing 50 μg of total protein were first digested with endoproteinases LysC. In the second and third digestion steps, trypsin and chymotrypsin were used (A, B). Peptide yields were determined by fluorometry (C), and the fractions were analyzed by LC-MS/MS (D, E). In the experiment with LysC the peptides generated by this enzyme were fractionated into four SAX fractions, whereas the tryptic peptides were separated into two SAX fractions. The peptides resulting from chymotryptic digestion were analyzed directly by LC-MS/MS.

isotopic labeling kits, may constrain sample size. To test the efficiency of multiple enzyme digestion for deeper proteome analysis, we analyzed samples of HeLa lysate containing 50 μg of total protein. These samples were consecutively digested with three enzymes: LysC, trypsin, and chymotrypsin (the last enzyme was added because we noticed that usable material still remained after trypsin digestion). In a control experiment the same sample was digested only with trypsin, and for consistency the retentate was cleaved with chymotrypsin (Figure 5A,B). The peptide yields obtained with trypsin and LysC were similar to those obtained in the above experiments, whereas chymotrypsin yielded an additional 1.5−2.5 μg of peptide material (Figure 5C). The peptide amounts obtained in these experiments were sufficiently high to allow prefractionation of the peptides before LC separation. For this we used a recently developed strong anion exchange (SAX) based technique in a pipet-tip-based format for fractionation into six samples,16 a method that has already been used in several studies.13,21−25 Here we employed the SAX approach directly for analysis of the reference tryptic digest but modified it for the LysC and

following tryptic digests to achieve the most efficient combination of sample amount and measurement time. The LysC fraction was separated into four SAX fractions and the postLysC tryptic digest into two fractions. The chymotryptic digests, which additionally yielded 1.5−2.5 μg of peptides, were analyzed directly without SAX fractionation. Comparison of the MED approach with standard tryptic digestion revealed an increase of 10% in the number of identified peptides and proteins (Figure 5D,E and Supporting Information Table 5). Although this improvement is not as high as in the experiments carried out with limited amounts of sample, we note that it was achieved without any additional resources such as mass spectrometric measuring time. 2635



Article

Figure 6. Identification of phosphorylation sites in (B) liver and (C) brain lysates. Mouse liver and brain lysates containing 2.2 and 1.6 mg of total protein, respectively, were consecutively digested with LysC and trypsin. Phosphopeptides were enriched on TiO2 beads and were analyzed by LCMS/MS on a Velos Orbitrap instrument. For comparison identical samples were digested with trypsin only and were analyzed by LC-MS/MS over identical time as the doubly digested sample. The liver sample was analyzed in triplicate whereas the brain lysate was assayed twice. (D−F), Examples of phosphorylation sites identified in brain proteins(D) potassium voltage-gated channel Q2, (E) electroneutral potassium-chloride cotransporter 2, and (F) Piccolousing MED FASP (LysC + trypsin) and tryptic digestion only. The heights of the bars show the peptide score values.

Two-Step Digestion Increases the Depth of Phosphoanalysis. In a recently published work, Tran et al.26 demonstrated clear advantages of a secondary digestion of large tryptic peptides for the identification proteins and phosphorylation sites. Their approach involved a size-exclusion chromatography step after tryptic digestion of proteins and a secondary digestion of aliquots of larger peptides with chymotrypsin, GluC, AspN, and formic acid. Adopting this in-depth strategy increased the number of identified phosphosites by 30%. Inspired by this study, we tested our MED FASP approach for the identification of phosphorylation sites. We analyzed mouse liver and brain lysates by two-step

digestion with LysC and trypsin and compared it to standard tryptic digestion (Figure 6 and Supporting Information Table 6). We observed an about 2-fold increase in the number of identified phosphorylation sites when we compared the twostep LysC/trypsin and single step trypsin digestion. In contrast, even combination of the data obtained in two independent single trypsin digestion control experiments increased the identified phosphoproteome by only 10−15%. These results clearly demonstrate the advantages of the multiple digestion approach over using a single enzyme for the identification of posttranslational modifications. 2636



Article

(8) Gatlin, C. L.; Eng, J. K.; Cross, S. T.; Detter, J. C.; Yates, J. R. 3rd. Anal. Chem. 2000, 72, 757−763. (9) Yates, J. R. 3rd; McCormack, A. L.; Schieltz, D.; Carmack, E.; Link, A. J. Protein Chem. 1997, 16, 495−497. (10) Ethier, M.; Hou, W.; Duewel, H. S.; Figeys, D. J. Proteome Res. 2006, 5, 2754−2759. (11) Vasilescu, J.; Zweitzig, D. R.; Denis, N. J.; Smith, J. C.; Ethier, M.; Haines, D. S.; Figeys, D. J. Proteome Res. 2007, 6, 298−305. (12) Manza, L. L.; Stamer, S. L.; Ham, A. J.; Codreanu, S. G.; Liebler, D. C. Proteomics 2005, 5, 1742−1745. (13) Wisniewski, J. R.; Nagaraj, N.; Zougman, A.; Gnad, F.; Mann, M. J. Proteome Res. 2010, 9, 3280−3289. (14) Wisniewski, J. R.; Zielinska, D. F.; Mann, M. Anal. Biochem. 2011, 410, 307−309. (15) Nielsen, P. A.; Olsen, J. V.; Podtelejnikov, A. V.; Andersen, J. R.; Mann, M.; Wisniewski, J. R. Mol. Cell. Proteomics 2005, 4, 402−408. (16) Wisniewski, J. R.; Zougman, A.; Mann, M. J. Proteome Res. 2009, 8, 5674−5678. (17) Larsen, M. R.; Thingholm, T. E.; Jensen, O. N.; Roepstorff, P.; Jorgensen, T. J. Mol. Cell. Proteomics 2005, 4, 873−886. (18) Olsen, J. V.; Macek, B.; Lange, O.; Makarov, A.; Horning, S.; Mann, M. Nat. Methods 2007, 4, 709−712. (19) Cox, J.; Mann, M. Nat. Biotechnol. 2008, 26, 1367−1372. (20) Cox, J.; Neuhauser, N.; Michalski, A.; Scheltema, R. A.; Olsen, J. V.; Mann, M. J. Proteome Res. 2011, 10, 1794−1805. (21) Weekes, M. P.; Antrobus, R.; Lill, J. R.; Duncan, L. M.; Hor, S.; Lehner, P. J. J. Biomol. Tech. 2010, 21, 108−115. (22) Zielinska, D. F.; Gnad, F.; Wisniewski, J. R.; Mann, M. Cell 2010, 141, 897−907. (23) Walther, D. M.; Mann, M. Mol. Cell. Proteomics 2010, 10, No. M110 004523. (24) Wisniewski, J. R.; Ostasiewicz, P.; Mann, M. J. Proteome Res. 2011, 10, 3040−3049. (25) Ostasiewicz, P.; Zielinska, D. F.; Mann, M.; Wisniewski, J. R. J. Proteome Res. 2010, 9, 3688−3700. (26) Tran, B. Q.; Hernandez, C.; Waridel, P.; Potts, A.; Barblan, J.; Lisacek, F.; Quadroni, M. J. Proteome Res. 2011, 10, 800−811.

Figure 6D−F shows three examples of sites detected on brain proteins: the potassium voltage-gated channel Q2, the electroneutral potassium-chloride cotransporter 2, and the protein piccolo. The majority of the sites were identified by MED FASP, whereas only a few were identified by tryptic digestion alone. Interestingly, for each protein more than 50% of the detected sites match those reported in our previous large-scale study of the brain phosphoproteome.13 The plots in Figure 5 also show that in most cases where a phosphopeptide was identified by MED FASP and by tryptic digestion, the MED FASP procedure allowed identification with higher scores than the tryptic digestion alone (Supporting Information Table 4). This finding stresses an additional advantage of using MED FASP, namely, that it not only increases the number of identified phosphopeptides but also provides more confidence in the data.

■

CONCLUSION Although the advantages of digestion of the same sample with different enzymes have long been recognized,1−4 this strategy has so far not become widespread. It seems that the main reasons for this were the needs for more sample and time of analysis. In this work, we have developed an alternative protocol overcoming these limitations by ultrafiltration-based separation of orthogonal populations of peptides. Using similar sample amounts and measuring times, the MED-FASP approach allows identification of peptides and proteins and posttranslational modification sites to a depth that was not achievable by single enzyme digestion applied to the same amount of sample.

■

ASSOCIATED CONTENT

S Supporting Information *

Six tables as described in the text (xls). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected] or [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Max Planck Society for the Advancement of Science, from the European Commission’s seventh Framework Program (grant agreement HEALTHF4-2008-201648/PROSPECTS), and the Munich Center for Integrated Protein Science (CIPSM).

■

REFERENCES

(1) Gauci, S.; Helbig, A. O.; Slijper, M.; Krijgsveld, J.; Heck, A. J.; Mohammed, S. Anal. Chem. 2009, 81, 4493−4501. (2) Swaney, D. L.; Wenger, C. D.; Coon, J. J. J. Proteome Res. 2010, 9, 1323−1329. (3) Choudhary, G.; Wu, S. L.; Shieh, P.; Hancock, W. S. J. Proteome Res. 2003, 2, 59−67. (4) Wisniewski, J. R.; Zougman, A.; Nagaraj, N.; Mann, M. Nat. Methods 2009, 6, 359−362. (5) Zhou, H.; Hou, W.; Lambert, J. P.; Figeys, D. Anal. Bioanal. Chem. 2010, 397, 3421−3430. (6) Yates, J. R. 3rd; Link, A. J.; Schieltz, D. Methods Mol. Biol. 2000, 146, 17−26. (7) Shevchenko, A.; Wilm, M.; Vorm, O.; Mann, M. Anal. Chem. 1996, 68, 850−858. 2637


Consecutive Proteolytic Digestion in an Enzyme Reactor Increases

Recommend Documents