Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Analysis of proteins isolated from dissolvable polyacrylamide gels using mass spectrometry Ramanowski, Peter 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2006-0296.pdf [ 6.74MB ]
JSON: 831-1.0092576.json
JSON-LD: 831-1.0092576-ld.json
RDF/XML (Pretty): 831-1.0092576-rdf.xml
RDF/JSON: 831-1.0092576-rdf.json
Turtle: 831-1.0092576-turtle.txt
N-Triples: 831-1.0092576-rdf-ntriples.txt
Original Record: 831-1.0092576-source.json
Full Text

Full Text

ANALYSIS OF PROTEINS ISOLATED FROM DISSOLVABLE POLYACRYLAMIDE GELS USING MASS SPECTROMETRY by Peter Romanowski B .Sc , The University of British Columbia, 2001 A thesis submitted in partial fulfillment of the requirements for the degree of M A S T E R OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Experimental Medicine) THE UNIVERSITY OF BRITISH C O L U M B I A April 2006 © Peter Romanowski, 2006 ABSTRACT Polyacrylamide gel electrophoresis is a simple yet powerful means of resolving complex mixtures of proteins and is used across many fields of research, including proteomics. The in-gel digestion has become the staple tool for identification of proteins run on SDS-PAGE gels but has unfortunate shortcomings such as unpredictable sequence coverage and the obtain precise intact protein masses. Other methods such as electro-eliition and passive diffusion evade these limitations by recovering intact protein but have difficulties isolating protein of high molecular weight. In this study, the use of a dissolvable SDS-PAGE gel for the isolation of intact proteins and subsequent analysis by mass spectrometry was evaluated. The resolutions of standard bis-acrylamide (BIS) crosslinked gels were compared to dissolvable gels crosslinked with bis-acrylylcystamine (BAC) to show virtually identical resolving capabilities. The model protein a-casein was successfully extracted from B A C -crosslinked. gels along with several other standard proteins. Accurate intact mass measurements and specific enzymatic processing were done to show a mass shift corresponding to 8 phosphorylation sites. In addition, gel-extracted protein was digested in solution with several enzymes, including trypsin, to demonstrate higher peptide recoveries and sequence coverage compared to the widely used in-gel digestion approach. In light of these additional features which are not available with standard gels, such as the ability to obtain precise protein masses and increased sequence coverage when digested, BAC-crosslinked gels may well become a reasonable alternative for proteomics analyses. i i TABLE OF CONTENTS A B S T R A C T ii T A B L E OF CONTENTS i i i LIST OF TABLES v LIST OF FIGURES vi ABBREVIATIONS vii A C K N O W L E D G E M E N T S ix CHAPTER 1: INTRODUCTION 1 1.1 Current Methods in Proteomics 1 1.1.1 Proteomics 1 1.1.2 2-D Gel Electrophoresis 4 1.1.3 In-gel Digestion 5 1.1.4 Protein Identification 8 1.1.5 The P T M Approach and MALDI-TOF MS 9 1.1.6 The LC-MS/MS Approach and the QTrap 11 1.1.7 Post-Translational Modifications and Limitations of Current Methods. ..15 1.2 Protein Isolation from SDS-PAGE Gels 19 1.2.1 Methods for Protein Isolation 19 1.2.2 Dissolvable Polyacrylamide Crosslinkers 21 1.2.3 Bis-acrylylcystamine and its Applications 22 1.2.4 Proteomic Applications of B A C Gels 24 CHAPTER 2: M A T E R I A L S A N D METHODS 26 2.1 Materials 2.1.1 BAC-Crosslinked Gels 26 2.1.2 QTrap Analysis of HS A 26 2.2 Methods 27 2.2.1 BAC-Crosslinked Gels... 27 Gel Casting/Electrophoresis : 27 Protein Isolation 28 Enzymatic Processing 31 In-Solution Digestion 31 In-Gel Digestion 32 ii i Coomassie Staining 34 Mass Spectrometry 34 2.2.2 QTrap Analysis of HS A 36 Chemical Acetylation 36 Liquid Chromatography 36 Mass Spectrometry 37 Experimental Variables 38 CHAPTER 3: M A S S S P E C T R O M E T R Y A N A L Y S I S OF INTACT PROTEINS ISOLATED F R O M DISSOLVABLE SDS-PAGE GELS 39 3.1 Rationale and Hypothesis 39 3.2 Results and Discussion 40 3.2.1 T E M E D / B A C Relationship 40 3.2.2 Pre-Electrophoresis 42 3.2.3 Resolution Comparison 46 3.2.4 Electrophoresis with Reducing Conditions 48 3.2.5 Precipitation of Dissolved Polyacrylamide with Organic Solvents 50 3.2.6 Extracted Proteins 53 3.2.7 Digestions 59 3.2.8 Concluding Remarks 64 CHAPTER 4: SYSTEMATIC E V A L U A T I O N OF D A T A ACQUISITION CRITERIA FOR COMPREHENSIVE ANALYSIS OF MODIFIED PROTEINS ON A HYBRID TRIPLE QUADRUPOLE/LINEAR ION TRAP INSTRUMENT 65 4.1 Rationale and Hypothesis 65 4.2 Results and Discussion 67 4.2.1 Acetylation of Human Serum Albumin 67 4.2.2 Peptide Sequencing and the Q-Trap 70 4.2.3 Combined Survey Scans 71 4.2.4 PREC Sensitivity .' 75 4.2.5 E M C Detuning 78 4.2.6 EMS and Dynamic Fil l Time 85 4.2.7 Combining Optimized EMS with PREC : 89 4.2.8 Concluding Remarks 91 CHAPTER 5: G E N E R A L CONCLUSIONS 93 5.1 Summary 93 5.2 Future Directions 94 REFERENCES 96 iv LIST OF TABLES Table 1 Comparison of detected peptides from trypsin, Glu-C, and chymotrypsin digestions of standard, BAC-extracted, and in-gel Cf-casein 62 Table 2 EMS DFT Trap Fi l l Times and Intensity 87 v LIST OF FIGURES Figure 1.1 In-Gel Digestion Procedure 7 Figure 1.2 MALDI-TOF and QTrap 12 Figure 1.3 Scan types on the QTrap Instrument 16 Figure 1.4 Dissolvable Polyacrylamide Crosslinkers 23 Figure 2.1 Protein Isolation from a Dissolvable Gel 29 Figure 3.1 Balancing B A C and T E M E D Concentrations 43 Figure 3.2 Importance of Pre-electrophoresis 45 Figure 3.3 Resolution Comparison Between BIS and B A C Gels 47 Figure 3.4 Reducing SDS-PAGE with B A C Gels :...49 Figure 3.5 Precipitation of Dissolved Polyacrylamide and Protein Solubility 52 Figure 3.6 Purification of Myoglobin 55 Figure 3.7 Purification of Fetuin 56 Figure 3.8 Intact Mass Measurements of Unprocessed and Processed a-casein 57 Figure 3.9 Gel-Extracted and In-Gel Tryptic Digestions 60 Figure 4.1 Sequence of Chemically Acetylated Human Serum Albumin 69 Figure 4.2 Q-Trap Workflow 72 Figure 4.3 Combining the survey scans: E M C + PREC 74 Figure 4.4 Modified Peptide Detection and Digest Load. 77 Figure 4.5 Increased PREC Dwell Time 79 Figure 4.6 Modified and Unmodified Peptides 81 Figure 4.7 E M C Trap Fi l l Time and Signal Intensity 84 Figure 4.8 Combining the survey scans: PREC + Optimized EMS with DFT 90 vi A B B R E V I A T I O N S p-ME P-mercaptoethanol APS ammonium persulfate B A C N, N '-bis-acrylylcystamine BIS N,N '-methylene bisacrylamide BSA bovine serum albumin CIP calf intestinal alkaline phosphatase cps counts per second Da Daltons DATD N, N '-diallyltartardiamide DFT dynamic fill time D H E B A N,N'-(l ,2-dihydroxyethylene)bisacrylamide D N A deoxyribonucleic acid DTT dithiothreitol EDA ethylene diacrylate E M C enhanced multiply charged EMS enhanced mass spectrometry EPI enhanced product ion ER enhanced resolution ESI electrospray ionization HC1 hydrochloric acid HFIP 1,1,1,3,3,3 -hexafluoroisopropanol HPLC high performance liquid chromatography HSA human serum albumin IDA information dependent acquisition IMAC immobilized metal affinity chromatography IPG immobilized pH gradient LC liquid chromatography LIT linear ion trap m/z mass to charge ratio M A L D I matrix assisted laser desorption ionization MS mass spectrometry MS/MS tandem mass spectrometry MudPIT multidimensional protein identification technology N L neutral loss N M R nuclear magnetic resonance PAGE polyacrylamide gel electrophoresis pi isoelectric point PMF peptide mass fingerprint PREC precursor ion P T M post translational modification RNA ribonucleic acid RP reverse phase RT room temperature vii S/N signal to noise ratio SCX strong cation exchange SDS sodium dodecyl sulfate TAP tandem affinity purification TEMED N, N,N',N '-tetramethylethylenediamine Th Thomson TOF time of flight Tris tris(hydroxymethyl) aminomethane U V ultraviolet XIC extracted ion chromatogram viii ACKNOWLEDGEMENTS I would like to thank Dr. Juergen Kast, my supervisor, for all of his guidance, and financial support. In addition, I thank Jason Rogalski for all his help with mass spectrometry and Dr. Peter Schubert for his advice, and help with samples which were run on the dissolvable polyacrylamide gels. ix CHAPTER 1: INTRODUCTION 1.1 Current Methods in Proteomics 1.1.1 Proteomics Proteomics is a discipline devoted to studying the complement of proteins, or proteome, of a given cell at a particular time and state [O'Farrell, 1975; Celis et al, 1996]. The field of proteomics can be divided into three main areas of research: protein microcharacterization, protein-protein interactions, and differential display [Pandey et al, 2000]. Protein micro-characterization is concerned with the large-scale identification of proteins.and their post-translational modifications or splice variants [Seo et al, 2004]. Post-translational modifications are important aspects of protein characterization since they govern protein function by determining activity [Harris et al, 2005], stability [Calatayud et al, 2005], and localization [Dauphinee et al, 2005], thus making modifications critical pieces of information in proteomics. Protein phosphorylation, for instance, is the major means of information transmission within the cell. Kinases are responsible for regulation of most of the signals from cell-surface receptors through phosphorylation of their substrates [Mann et al., 2001]. A viable approach to studying many post-translational modifications including serine/threonine/tyrosine phosphorylation is to purify specifically modified proteins with antibodies directed against the modification. Currently, there are several commercially-available anti-phosphoserine, anti-phosphothreonine, anti-phosphotyfosine and a host of other modification-specific antibodies used for such experiments. These antibodies are used to 1 purify specific modifications by immunoprecipitation followed by SDS-PAGE to resolve the proteins. Bands from the gel are excised and analyzed by mass spectrometry (MS) to identify the proteins [Pandey et al, 2000; Pandey et al, 2000]. A n alternative approach is to resolve cell lysates by 2-dimensional (2D) gel electrophoresis, perform a Western blot using a modification-specific antibody to identify all the spots containing modified proteins and then use MS analysis to identify the proteins [Soskic et al, 1999]. While mass spectrometry is the core technique in proteomics, there is, however, an issue with using it to identify modification sites on proteins: proteins of interest must be enriched significantly to provide enough material for analysis [Mann et al., 2001]. The second area is focused on correlating identification and analysis to the function of genes and proteins. Since most proteins occur in protein complexes, understanding the function of the protein within the cell necessitates identification of its interacting partners [Pawson et al., 1997]. Identification of protein-protein interactions is possible using techniques such as yeast-2-hybrid [Davies et al, 2004] or in-vivo formaldehyde crosslinking [Vasilescu et al, 2004; Rappsilber et al, 2000]. A simpler approach to purifying an entire complex associated with a protein of interest is using an antibody against that protein [Witke et al, 1998]. A similar but more common approach, however, is to fuse an epitope tag, such as F L A G , H A or Myc tag, to the protein of interest. The protein is then transfected into the cell, over-expressed and immunoprecipitated with an antibody directed against the tag [Lichty et al, 2005; Figeys et al, 2001]. Once immunoprecipitated, the entire complex is competitively eluted by a soluble peptide encoded for the epitope tag used [Mann et al., 2001]. The complex is then resolved by SDS-PAGE and the resulting bands are excised and analyzed by MS. 2 This approach is necessary for all but the most abundant proteins to ensure sufficient material is obtained for analysis, or when there is no commercially available antibody that would be specific, for the protein of interest. A concern for using this type of approach is antibody specificity as a properly controlled set of experiments is necessary to separate non-specific background proteins from true interacting partners. Addressing this issue has been the focus of purification systems such as the Tandem Affinity Purification (TAP) method where a dual epitope tag approach is used [Puig et al, 2001; Rigautef.a/., 1999]. Finally, differential display studies compare protein expression of one particular cell state to another by attempting to resolve the cell lysates on separate 2D gels [Ostergaard et al, 1999; Page et al, 1999]. These gels are then visually compared for pattern differences between two cell states, such as cancerous vs. normal tissues. Protein spots which become more intense, fainter, or disappear relative to a control or untreated cell state are all candidates for further study and are analyzed by MS for identification. The objective is to identify all proteins which change in terms of expression or modification status relative to a background of proteins which remains unaffected by the changed conditions. Drawbacks to this approach include: the failure to resolve most proteins greater than 150kDa, the inability to separate most membrane proteins, the failure to consistently detect more than 1000 proteins by MS and the inability to resolve extremely basic and acidic proteins [Mann et al, 2001]. As an alternative to analyzing protein spots from a 2-D gel, other techniques such as 2-dimensional liquid chromatography (2-DLC) and Multidimensional Protein Identification Technology (MudPIT) are also available. With 2-DLC, a digested tissue 3 extract is separated using two columns in series; typically one strong cation exchange (SCX) and one reverse phase (RP) column. Fractions collected from the first column are then individually applied to the RP column for separation [Opiteck et al, 1997; Davis et al, 2001]. Peptides are then analyzed either by a U V detector or using electrospray ionization mass spectrometry (ESI-MS). An analogous method to 2-DLC is MudPIT which uses a single biphasic column containing the same SCX and RP packing materials. An advantage of packing both in the same column is that valves and dead volumes common to all 2DLC systems are eliminated [Washburn et al, 2001]. After the sample is loaded, a series of salt steps elute sample from the S C X half of the column while a ramped concentration of organic solvent between each step elutes peptides which have moved to the RP half. Peptides eluted from the RP half of the column are analyzed directly by tandem mass spectrometry. A limitation of this technique is sample throughput since custom column packing and sample loading are entirely manual. Furthermore, each experiment may require long instrument times depending on the number of salt steps used [Link et al, 1999]. Unfortunately, all multidimensional separation techniques suffer from the common limitation of sample detection, mainly by mass spectrometry [Liu et al, 2002]. 1.1.2 2-D Gel Electrophoresis Two-dimensional gel electrophoresis has become the tool of choice for many proteomics-based studies. Its ability to reproducibly display hundreds of proteins on a single gel makes it a powerful technique for examining and comparing different proteomes [Thiede et al, 2005]. The two dimensions of separation and the resolution of 4 each dimension give this technique the ability to display so many proteins on a single gel. The first dimension is isoelectric focusing on gel strips which contain an immobilized pH gradient (IPG). The pH gradient allows for protein separation based on the unique isoelectric point (pi) of each protein when a voltage is applied step-wise across the strip. One limitation is that proteins with extreme pis (ie., very acidic or very basic) cannot be separated during this step [Fichmann et al., 1999]. The IPG strip is then loaded onto a regular bis-acrylamide SDS-PAGE gel to separate the proteins again, but this time based on their molecular weight. For this, SDS-PAGE is used as the second dimension [Klose et al, 1995]. In contrast to 1-D gels, where proteins are only separated based on their molecular weight and appear as horizontal bands, 2-D gels present proteins as spots. Having this kind of superb resolution is necessary; however, a sensitive and rapid, preferably automatable, analytical method to identify and characterize the proteins found within the spots is also necessary [Bunai et al., 2005]. Identification of the spots is done using a commonly-used technique known as in-gel digestion where gel-separated proteins are digested into peptides by sequence-specific proteases such as trypsin, extracted from the gel and analyzed by mass spectrometry. 1.1.3 In-gel Digestion In-gel digestion of protein-containing SDS-PAGE bands or spots has become the essential tool for the proteomics researcher in terms of the protein identification (Fig. 1.1). After staining with either standard Coomassie Brilliant Blue (R-250) or silver, spots or bands can be excised robotically or manually for identification of the proteins contained within. Many proteins contain disulfide bridges between two cysteine side 5 chains to serve as stabilizers of tertiary structure. For proper digestion, these linkages must be broken so that two peptides are not linked together during the identification [Shevchenko et al, 1996]. Databases used for identification of digested proteins do not account for peptides still linked by disulfides. To ensure disulfides are broken, reducing agents such as dithiothreitol or (3-mercaptoethanol are used to return the individual cysteine side chains back to reduced thiols. While it is true that these same reducing agents are used during electrophoresis in the sample loading buffer and that disulfides should already be broken, the reduction step is done nevertheless. Since the reaction of disulfide reduction is a simple exchange mechanism of thiolate ions attacking and displacing other thiols, disulfide bond reformation may occur. In order to prevent such events, the reduced thiol groups are alkylated, or blocked, with iodoacetamide, iodoacetic acid or other thiol alkylating agent at pH 7-8 [Shevchenko et al, 2001]. Iodoacetamide reacts preferentially with free sulfhydryl groups, but alkylation is possible with histidine and methionine side chains as well as alkylation of lysine side chains with a higher pH [Boja et al, 2001]. Once all cysteines have been blocked, gel bands are incubated in a solution containing the protease of choice. Trypsin is most commonly used since every tryptic peptide, other than the C-terminal peptide, has a chargeable basic residue at its C-terminus and because the specificity is after K and R, peptides are statistically between 10-15 amino acids which are ideal for mass spectrometric analysis. After digestion is complete, peptides are extracted and the identity of the unknown protein is deduced through MS analyses [Corthals et al, 2000]. 6 Figure 1.1 In -Ge l Digestion Procedure. Workflow diagram for a typical in-gel digestion protocol used for identification of unknown proteins from SDS-PAGE bands. Cut out band(s) of interest Reduce protein disulfide bonds with DTT Alkylate cysteines with Iodoacetamide BIS-crosslinked SDS-PAGE gel Overnight digestion with trypsin Peptide Mass Fingerprint m + MALDI-TOF MS m/z Peptide extraction < Database search + Identification 7 1.1.4 Protein Identification Currently, there are two main analytical approaches for protein identification using peptides: peptide mass fingerprinting (PMF) [Henzel et al, 1993] or by peptide sequencing by liquid chromatography-coupled tandem mass spectrometry (LC-MS/MS). The PMF approach analyzes the eluted peptide mixture on a MALDI-TOF and compares the signals in the MS spectrum against the peptide masses provided by in-silico digestions of all protein sequences entered in the database [Jensen et al, 1997; Berndt et al, 1997]. The method is sensitive, fast and can be completely automatic so that hundreds or even thousands of samples can be analyzed and their identities deciphered in relatively short time. The caveat is that the protein submitted to this search procedure has to be contained in the database to begin with. Furthermore, i f one is searching for post-translational modifications, the peptide mixture must first contain the modified peptide and the modification must already be known in the database [Schmidt et al, 2003]. Thus, finding new post-translational modifications using this approach is not possible unless the peaks are manually inspected for differences corresponding to possible modifications. For instance, a shift of +80Da from a predicted peptide mass could be indicative of serine, threonine or tyrosine phosphorylation and even tyrosine sulfation [Kange et al, 2005; Gomez et al, 2002; Loughnan et al, 1998]. Other protein isoforms, such as products of alternative splicing or proteolysis, may lead to altered or truncated protein sequences and, theoretically, can also be detected with PMF. Unfortunately, these are even more difficult to confirm since the absence of peptides are by no means conclusive evidence for truncations. 8 The second approach relies on first resolving the peptide mixture into single peptides via liquid chromatography followed by analysis using MS fragmentation and sequencing - the so-called LC-MS/MS approach. Automated L C - M S / M S analysis acquires MS/MS spectra for hundreds to thousands of peptides in a single L C run. Fragmentation of peptides using tandem mass spectrometry provides unambiguous information of not only peptide sequence but also the location and nature of post-translational modifications and sequence variants [Liebler et al, 2002]. The analysis is carried out on a tandem mass spectrometer, such as a triple quadrupole [Gasthuys et al, 2004] or a QqTOF instrument [Warren et al, 2005]. Other instruments are also certainly possible, but these platforms currently predominate together with ion trap instruments [Lin et al, 2003]. Sequence information derived from several peptides is much more specific for identification of a protein than from a list of peptide masses, as in the PMF approach [Pandey et al, 2000]. Specific post-translational modification detection is also possible using this approach and requires no previous assignment in a database. Unfortunately, liquid-chromatography-based MS/MS sequencing is much more time consuming compared to PMF regardless of the HPLC gradient times used. 1.1.5 The P T M Approach and M A L D I - T O F M S Matrix Assisted Laser Desorption/Ionization (MALDI) is an ionization source used for mass spectrometric analysis of peptides [Gevaert et al, 2001], proteins [Bolbach, 2005], nucleic acids [Mauger et al, 2006], metabolites [Babu et al, 2006], and other small molecules [Riebeseel et al, 2002]. Its high sensitivity, high throughput and ease of operation make M A L D I a widely used MS analytical method for protein 9 identification since its discovery in 1988 [Karas et al, 1988; Tanaka et al., 1988]. In M A L D I , the sample is mixed into a solution of matrix and spotted onto a metal target which is then inserted into the mass spectrometer. The matrix is responsible for absorbing and transferring laser energy to the analyte in a manner which prevents sample decomposition. During this process, the analyte is ionized by proton transfer from the excited matrix molecules, most likely in the solid phase [Hillenkamp et al., 1991]. Unlike electrospray ionization (ESI) where multiply-charged ions are formed, M A L D I produces predominantly singly protonated species, however doubly- and triply-charged forms are also observed, as in the case for some proteins. Because the entire process of desorption and ionization is initiated by the pulsing of the laser, the M A L D I source is ideally combined with the Time of Flight (TOF) mass analyzer since the delayed extraction of ions after the laser pulse can start the time for ions traveling from the source to the detector [Fenselau, 1997]. Other mass analyzers including magnetic sectors [Kolli et al, 1997], ion cyclotron resonance [DeKeyser et al, 2006], ion traps [Von Seggem et al, 2003], and TOF-TOFs [Traub et al, 2006] have also been coupled to M A L D I sources, although MALDI-TOF is the most common combination. Different matrices and different modes of mass analysis are used for different samples. For instance, the matrix a-cyano-4-hydroxycinnamic acid is mainly used for peptides [Jagannadham et al, 2005] whereas sinapinic acid is commonly used for proteins [Schreiner et al, 1996]. As well, during TOF mass analysis, proteins travel directly to a detector (linear mode) in contrast to peptides which go through a reflectron field for enhanced resolution (reflectron mode). The paths proteins and peptides travel within a TOF tube are shown in Figure 1.2A. 10 1.1.6 The L C - M S / M S Approach and the QTrap Since the L C - M S / M S approach relies on obtaining sequence information for individual peptides, liquid chromatographic separation is used to resolve them within a complex digestion mixture. These peptides are then ionized using electrospray ionization and enter a tandem mass spectrometer such as a triple quadrupole instrument. These types of instruments are well suited for the selection of peptides and their fragmentation for sequence information since these processes are physically separated in space. A quadrupole is made of four parallel metal rods arranged as shown in Figure 1.2B. Voltages are applied to the rods which make them controlled completely by electronics. Since the quadrupoles are controlled electronically, many scanning options are available, making them a very flexible and versatile means of mass analysis. For mass analysis, ions travel in between the rods with their flight trajectories being affected by these applied voltages. A specific set of voltages applied to the rods allows for a fixed m/z ratio to pass through the quadrupole while all other m/z values will have unstable flight through the quadrupole and strike the rods. Thus, the quadrupole is able to act as a mass filter. Conversely, quadrupoles may act as "ion guides", allowing all ions to pass through without any specific m/z favouring. Arranging three of them in series (triple quadrupole) gives the operator many different scan types for targeted analysis. The first and third quadrupoles (Ql + Q3) are operated in one of the modes described above: fixed m/z scanning to allow for one ion to pass through or allow all ions to pass through regardless of their m/z ratio. Alternatively, quadrupoles may also perform step-wise scanning through a range of m/z ratios, allowing ions of a certain m/z to pass before moving to the next m/z. The second quadrupole (Q2) can be used for sample fragmentation when 11 Figure 1.2 MALDI-TOF and QTrap. Schematic diagrams of a MALDI-TOF (A) and the recently introduced hybrid triple quadrupole/linear ion trap (QTrap) (B) mass spectrometers. Trapping grids 12 sequence information is desired. This quadrupole is known as the "collision cell" since sample ions collide with neutral gas molecules, known as the collision gas, and this leads to fragmentation of the sample i f the energy of the ion is high enough. Traditionally, Q2 was either "empty" with no collision gas or "filled" with a higher pressure of collision gas where fragmentation would occur. This, however, is no longer true on modern collision cells since collision gas is always present at the same pressure. Fragmentation is now controlled by potential differences where higher voltages give ions more energy for collision and lead to fragmentation. Low potential energies do not cause ion fragmentation in Q2 due to insufficient kinetic energy [Sleno et al, 2004]. Triple quadrupoles are rugged and have a variety of scan types available, but are not as sensitive as, for instance, 3D ion trap instruments which can accumulate and store ions prior to scanning them out. Ion trap instruments work by trapping ions within a ring electrode which separates two hemispherical electrodes. Once ions are admitted into the trap, changes in the electrode voltages are used to mass-selectively eject ions through the hemispherical electrodes. Advantages of the 3D ion trap are that they are physically small, inexpensive, very sensitive and have MS" capabilities [Hopfgartner et al., 2004]. Unfortunately, they have poor trapping efficiencies (1-10%) and small trapping volumes, making ion storage capacity limited. Overfilling the trap with ions causes a phenomenon known as "space charge effects" where the electric fields of ions are distorted due to too many ions spaced too closely. Space charge effects result in deterioration of the mass spectrum and loss in dynamic response range [Hager, 2002]. The recently introduced QTrap instrument combines both the scanning variety and robustness of a triple quadrupole with the trapping functionality of an ion trap by 13 incorporating a quadrupole linear ion trap (Fig. 1.2B). The geometry of the quadrupole linear ion trap is analogous to a regular quadrupole except that ions can be accumulated. Fringe fields at the end of the quadrupole are used to mass-selectively eject the trapped ions through the end [Hopfgartner et al, 2004]. The linear ion trap has two main advantages over traditional 3D ion traps: it has higher trapping efficiencies and larger ion storage capacities. The third quadrupole (Q3) in the QTrap is able to function as a regular quadrupole or accumulate ions as a linear ion trap [Hager, 2002]. The trapping function in Q3 enables the QTrap to have even more possible scan types and gives superior sensitivity over conventional triple quadrupoles since trapping can be involved. Possible scan types on the QTrap include the "Enhanced M S " (EMS), the "Enhanced Multiply Charged" (EMC) the "Enhanced Product Ion" (EPI) and the traditional Precursor Ion (PREC) scan (Figure 1.3). When trapping is used, as in the cases of EMS, E M C , and EPI scans, this is designated by using the word "Enhanced". The EMS scan is used as a general MS scan with ions traveling directly to Q3 for trapping. Ions of all charge states are then scanned out of Q3 to produce single-MS spectra with increased signal intensities due to trapping (Fig. 1.3A). The E M C scan is used as a comprehensive MS survey scan which has the added capability of removing ions with low charge states (ie. singly charged) prior to scanning out the remainder of the predominantly multiply charged ions from Q3 (Fig. 1.3B). This is an effective way of removing singly charged impurities such as solvent clusters or polymers and select for multiply-charged peptides which give superior fragmentation. These qualities make it a good general survey scan for experiments involving peptide sequencing and subsequent protein identification. In terms of intensity, EMS scans are more intense than E M C scans 14 as all the ions are used for the spectrum. E M C scans, however, are superior to EMS scans in terms of signal: noise ratios since much of the noise is singly-charged and would be removed. The EPI scan selects one type of ion in Q l , fragments it in Q2 and accumulates the fragment ions in Q3 for increased sensitivity (Fig. 1.3D). The fragments detected from Q3 are recorded in a mass spectrum and software can be used to submit this information to a database search for protein identification [Hager et al, 2003]. Triple quadrupole-based instruments are ideal for post-translational modification (PTM) screening as they are able to fragment peptides and detect characteristic marker ions created by modified side chains or neutral mass shifts using precursor ion (PREC) or neutral loss (NL) scans, respectively. The decision between using one scan over the other is dependent on whether the modification under consideration produces a charged marker ion which can be detected by PREC (Fig. 1.3C) or an uncharged neutral loss for the N L scan. Stable modifications such as lysine acetylation or tyrosine phosphorylation are detected with the PREC scan for marker ions at 126.1Th [Kim et al, 2002] or 216.1 Th [Salek et al, 2003], respectively, whereas labile modifications such as serine or threonine phosphoylation can be deduced from a neutral loss of 80Da [Hogan et al, 2003; Huq et al, 2005]. Alternatively, serine and threonine phosphorylations can also be detected at -79 and -97 m/z using PREC scanning in negative mode [Williamson et al., 2006]. 1.1.7 Post-Translational Modifications and Limitations of Current Methods Post-translational modifications are reversible, covalent additions found on amino acid side chains which are responsible for controlling many features of proteins including 15 Figure 1.3 Scan types on the QTrap Instrument. Quadrupole (Q) settings for (A) Enhanced Multiply Charged (EMC), (B) Enhanced MS (EMS), (C) Precursor ion (PREC) and ( D) Enhanced Product Ion (EPI) scans. Quadrupoles with brackets indicate when trapping in Q3 is used. Scans involving fragmentation such as the PREC and EPI scans have a "filled" Q2. Q1 A Open Q1 B Open Qf Step-wise scan Q1 Fixed on one m/z Filled Q3 Trap ions & scan out multiply charged peptides ) Trap ions & scan out Q3 Fixed on a marker ion Q3 m " Trap fragments & scan out 16 activity [Wang et al, 2006; Calissano et al, 2005], interactions with other proteins or D N A [Kunz et al, 2005; Wunsch et al, 2005], lifespan [Oberst et al, 2005], and even their localization in subcellular compartments [Keller et al, 2005]. In signaling, for instance, kinase cascades are regulated by the addition and removal of phosphate groups [Cohen, 2000]. As well, ubiquitination of, for instance, cyclins marks them for destruction at defined points in the cell cycle [Tyers et al, 2000]. Moreover, gene expression is partially controlled by acetylation and methylation of lysine residues of histones binding to D N A [Allfrey et al, 1964]. When modified, the ionic interaction between the positively-charged lysine side chain and the negatively-charged D N A backbone is perturbed, exposing sections of the chromosome to the transcription machinery [DeLange et al, 1970]. Thus, these very minor changes to a protein's primary structure provide insight into protein function. Analysis of PTMs by mass spectrometry has become the tool of choice for detection and localization but is far from being a trivial or routine task. The type of analysis used must be custom tailored to the modification being studied and relies on factors such as its stability and, obviously, the expected mass shift associated with it. Proteins which are weakly expressed or which contain modified residues in sub-stoichiometric amounts complicate matters even more. Enrichment steps using antibody purifications targeting the protein [Wang et al, 2006] or modification of interest [Komatsu et al, 2003] are necessary steps in order to yield enough material for analysis. Alternatively, proteins bearing certain modifications can be purified via specialized methods such as Immobilized Metal Affinity Chromatography (IMAC) for phosphoproteins [Yip et al, 1985; Belew et al, 1987] or by lectin-based purifications for 17 carbohydrate-modified proteins [Cheng et al, 2000; Stubb et al, 2004]. Separating these often complex mixtures on a 1-D or 2-D SDS-PAGE gel to resolve individual proteins is commonly followed by identification with the widely used in-gel digestion approach [Yang et al, 2004; Sultan et al, 2004]. In order to confidently identify proteins run with SDS-PAGE, only few peptides are required. Unfortunately, this rarely includes information on post-translational modifications or truncations. This is a serious downfall and limitation since these are the pieces of information which are sought after in proteomics. The standard identification is, really, only the first step towards full characterization of a particular protein and only provides a minimum amount of information. The root of the problem is that only peptides are isolated from the gel and this recovery varies from protein to protein. The in-gel digestion approach is not a reliable means of guaranteeing that one will obtain the modification-containing peptide from a gel band. Moreover, the intact mass of the protein isoform is typically not known, other than from estimates of electrophoretic migration, which can be inaccurate [Garrigos et al, 1991; Deschamps et al, 1992]. For post-translational analysis of proteins, the only way to ensure that the modification is obtained from a sample within an SDS-PAGE band is to isolate the entire protein, for analysis. When the entire protein is extracted from the band, several other possible experiments may be done in order to gain insight into modification status and/or splice variants. Intact mass measurements would provide the exact mass of the protein isoform and may hint at possible modifications i f a standard mass is previously known. A more convincing approach to proving the existence of modifications is post-elution enzymatic processing specific for certain types of modifications. For instance, alkaline 18 phosphatases and glycosidases are specific for removing phosphorylations [Fernley et al., 1968] and certain types of glycosylations [Sundblad et al., 1988; Yurewicz et al., 1991], respectively. Thus, isolating the intact protein from a polyacrylamide gel would overcome some of the inherent shortcomings of the current in-gel digestion procedure. 1.2 Protein Isolation from SDS-PAGE Gels 1.2.1 Methods for Protein Isolation The inability to obtain intact mass measurements or to perform specific enzymatic processing of potentially modified proteins in SDS-PAGE bands are the main limitations of the widely-used in-gel digestion procedure. Certain methods attempt to circumvent some of these limitations by focusing on the isolation of the intact protein from SDS-PAGE gels. Once the protein is isolated, this enables a number of additional analyses which may be done in order to gain more insight into its specific isoform. Intact mass measurements, enzymatic processing for post-translational modification, and digestion with different enzymes for increased sequence coverage are all possible once the intact protein is purified. Electro-elution [Schuhmacher et al, 1996], direct desorption from ultrathin gels [Ogorzalek et al, 1997], electroblotting [Vestling et al, 1994], and passive diffusion with [Ehring et al, 1997] and without [Feick et al, 1990] ultra-sonication have been previously shown to isolate intact proteins from SDS-PAGE bands. Electroelution is a common approach for extraction of proteins from SDS-PAGE gels and works by simply applying a voltage across the gel slab or band so that the still 19 negatively-charged proteins are drawn to the anode. Many different types of electroelution apparatuses have been designed which can operate with [Bhaskar et al, 2000] and without [Radko et al, 2002; Chang et al, 2001] gel sectioning, however it is possible to construct a simple electroelution device in the lab [Bongertz et al, 1989]. Various parameters including acrylamide percentage, gel thickness, buffer used, and the duration of the applied voltage have been evaluated [Dunn, 2004]. There are many parameters which can be varied, especially the type of solvent system or buffer used which leads to many protocols with little consensus on a standardized procedure compatible with subsequent MS analysis. Studies involving electroelution can routinely isolate BSA from SDS-PAGE bands with varying degrees of recovery [Yefimov et al, 2000; Chang et al, 2000]. Typical electroelution times can span from lOmin to 8 hours, depending on the voltage used and the size of the protein. A drawback to this approach is that contaminants, such as SDS, which are incompatible with MS analyses are also extracted from the gel and must be removed. A widely used alternative to electroelution is passive diffusion, a technique which relies solely on proteins diffusing from SDS-PAGE bands into a surrounding solvent system optimized for protein solubility. Passive elution is much simpler than electroelution, requiring no special equipment. Elutions can be done with bands directly after electrophoresis using water, 1% SDS, 50mM Tris + 200mM glycine (pH 8.0), and also in the M A L D I matrix a-cyano-4-hydroxycinnimic acid (20mM, pH 9.0) [Jorgensen et al, 2004]. However, sometimes more complicated solvent mixtures are employed such as formic acid/acetonitrile/isopropanol/water (50/25/15/10, v/v/v/v) [Ehring et al, 1997] or formic acid/water/isopropanol (1:3:2, v/v/v) [Cohen et al, 1997]. The duration 20 of passive elution varies from lOmin to, again, several hours and sometimes uses increased temperature and sonication to accelerate and maximize recovery [Ehring et al., 1997; Jorgensen et al, 2004]. Some drawbacks include a mass range limited to approximately 50kDa [Jorgensen et al, 2004; Claverol et al, 2003] and, for protocols using formic acid in the solvent mixture, protein adducts caused by formylation of serine and threonine side chains [Goodlet et al, 1990]. These systems should be used with caution if meaningful MS analyses are the ultimate goal. 1.2.2 Dissolvable Polyacrylamide Crosslinkers Dissolvable polyacrylamide gels are of value whenever the recovery of macromolecules from SDS-PAGE bands is desired [Faulkner et al, 1982; Flyer et al, 1982]. Being able to couple the excellent resolution of SDS-PAGE with the ability to extract and analyse the proteins contained within is a desirable feature which opens many other options. There are several cleavable polyacrylamide crosslinkers which were introduced during the mid-1960's up to the mid 1970's. This era saw the creation of several unique crosslinkers, each capable of causing gel solubilization under various conditions. The first of them was ethylene diacrylate (EDA, Fig. 1.4B) which contains ester bonds cleavable in alkaline media [Choules et al, 1965]. Next came N,N'-diallyltartardiamide (DATD, Fig. 1.4D) [Anker, 1970] and N,N'-(\,2-dihydroxyethylene)bisacrylamide (DHEBA, Fig. 1.4C) [O'Connell et al, 1976] which both contained 1,2-glycol structures breakable in the presence of lOmM periodic acid. D H E B A also contains base-cleavable amido-methylol bonds. The time and temperature needed for dissolution of gels containing these crosslinks varied from 30 min at room 21 temperature for D A T D to 12 hours at 50°C for D H E B A in lOmM periodic acid. Surprisingly, there are even reports of dissolving gels crosslinked with standard N,N'-methylene-bisacrylamide (BIS, Fig. 1.4A) under rather extreme conditions: 30% hydrogen peroxide at 37 or 50°C [Young et al, 1965; Goodman et al, 1971]. The fact that none of the above-mentioned dissolvable gel crosslinkers are in widespread use for extraction of macromolecules contained within SDS-PAGE gels is clear. Many macromolecules are, unfortunately, altered under the conditions necessary for cleavage of these crosslinkers and cannot be recovered undamaged [Hansen, 1980]. A crosslinker which is breakable under relatively benign conditions and leaves the analyte free of non-specific modification would certainly be a valuable asset for recovery of proteins and other molecules resolved with SDS-PAGE. 1.2.3 Bis-acrylylcystamine and its Applications The most recently introduced cleavable cross-linker, bis-acrylylcystamine (BAC, Fig. 1.4E), contains a disulfide linkage that is susceptible to thiol-reducing agents such as p-mercaptoethanol or dithiothreitol (DTT) [Hansen, 1976]. Created by Dr. J.N. Hansen in 1976, the B A C crosslinker was first used in a 3.3% polyacrylamide gel for recovery of RNA [Hansen, 1976] and later, D N A [Hansen 1981]. Since the percentage of acrylamide used for nucleic acids is relatively low, allowing for larger pore sizes, such conditions are unsuitable for resolving protein mixtures. The first account of protein isolation used gels incorporating the B A C crosslinker and acrylamide concentrations of up to 17% to successfully resolve histones in acid-urea-22 Figure 1.4 Dissolvable Polyacrylamide Crosslinkers. Chemical structures of: (A) TV,/V'-methylene bisacrylamide (BIS); (B) Ethylene diacrylate (EDA); (C) N,N'-(\,2-dihydroxyethylene)bisacrylamide (DHEBA); (D) A^N'-diallyltartardiamide (DATD); (E) N,N'-bis-acrylylcystamine (BAC). OH O O OH OH O O O 23 Triton gels [Faulkner et al, 1982]. The next example was the isolation of an 80kDa protein Simian Virus 40 Large T Antigen, which was resolved and purified from a 7.5% polyacrylamide gel [Flyer et al, 1982]. Later on, native pyridine nucleotide-linked dehydrogenases were resolved on a BAC-crosslinked gel and detected via an enzymatic activity stain using a substrate and cofactor. The enzyme was excised and purified successfully from a BAC-crosslinked band [Seymour et al, 1989], Polyacrylamide crosslinked with B A C has even been applied directly in medicine to create a bioartificial pancreas for treatment of insulin-dependent diabetes. The polymerized gel was used to create a hydrogel to mimic the islets of Langerhans - a special group of cells in the pancreas designed to secrete hormones such as insulin [Reckard et al, 1974; Boschero et al, 1971]. The hydrogels could then be liquefied to release insulin with reducing agents which is analogous to protein isolation from polyacrylamide gels using B A C [Hisano et al, 1998]. The use of disulfide-reducing agents with proteins is a standard, highly specific, and gentle procedure which leaves the analyte free of non-specific modification [Janusz et al, 1986; Grazu et al, 2003]. For this reason, we chose to investigate the applicability of the B A C cross-linker for gel electrophoresis and subsequent mass spectrometric analyses. 1.2.4 Proteomic Applications of B A C Gels Every area of proteomics, from protein micro-characterization to protein-protein interactions to differential display studies, utilizes either one-dimensional or two-dimensional SDS-PAGE separations to resolve protein mixtures for further mass 24 spectrometric analyses [Soskic et al., 1999; Puig et al., 2001; Ostergaard et al., 1999]. Polyacrylamide gels cross-linked with B A C have proven to be useful where the recovery of undigested biological material is desired after electrophoretic separation [Hansen, 1976; Hansen, 1981; Faulkner et al, 1982; Flyer et al, 1982; Seymour et al, 1989]. Since the three areas of proteomics are focused on the analysis of proteins and even their interaction partners by mass spectrometry, this type of dissolvable gel system may provide advantages for identification and characterization over the widely-used in-gel digestion approach. For protein micro-characterization, where the goal is large-scale identification of proteins along with characterization of all modification sites, recovery of intact proteins from SDS-PAGE gels would provide additional analytical options for modification study. The number phosphorylations, for instance, could be calculated by measuring the difference in intact weight of a protein before and after in-vitro phosphatase treatment. In studies involving protein-protein interactions, bait proteins and their interaction partners could be isolated from a dissolvable gel for intact mass measurements and digestion with a variety of proteases for increased sequence coverage. For differential display experiments using 2-D gel electrophoresis, individual protein isoforms could be recovered from the gel as separate spots. The isolated isoforms could be identified and characterized for differences in modification status and even splice forms using MS. The potential applications of dissolvable gels incorporating the B A C crosslinker are numerous and could be useful in all areas of proteomics. Thus we chose to investigate its potential for various proteomic applications and its amenability to mass spectrometric analyses. 25 C H A P T E R 2: M A T E R I A L S A N D M E T H O D S 2.1 Materials 2.1.1 BAC-Crossl inked Gels A,JV'-bis-acrylylcystamine (BAC) . crosslinker, N.N.N'.N'-tetramethylethylenediamine (TEMED), ammonium persulfate (APS), 1,1,1,3,3,3-hexafluoroisopropanol, P-mercaptoethanol (P-ME), dithiothreitol (DTT) and iodoacetamide were purchased from Sigma-Aldrich (St. Louis, M O , USA). Solvents including acetone, methanol, acetonitrile, glacial acetic acid, hydrochloric acid, formic acid, and isopropanol, were obtained from Fisher Scientific (Hampton, N H , USA). Acrylamide powder, 30% acrylamide/A^-A^'-methylene-bisacrylamide solution (37:1, 2.6% crosslinker), tris(hydroxymethyl)-aminomethane (Tris), Coomassie R-250, and sodium dodecyl sulfate (SDS) were obtained from Biorad (Hercules, C A , USA). Calf Intestinal Alkaline Phosphatase (CIP) along with the appropriate NEBuffer3 solution was purchased from New England Biolabs (Ipswich, M A , USA). Enzymes used for digestions such as sequencing grade modified trypsin (Bos taurus), endoproteinase Glu-C (Staphylococcus aureus) and chymotrypsin (Bos taurus), were obtained from Roche Diagnostics (Mannheim, Germany). 2.1.2 QTrap Analysis of H S A Human Serum Albumin (HSA), sodium borate, sodium hydroxide, and urea were purchased from Sigma-Aldrich (St. Louis, M O , USA). Markl2 Wide Range unstained 26 standard was obtained from Invitrogen Life Sciences (Carlsbad, C A , USA). Sequencing Grade Modified Trypsin used for all digestions was purchased from Promega (Madison, WI, USA). A l l solvents used for sample preparation or mass spectrometry such as acetone, acetonitrile, acetic anhydride and formic acid were purchased from Fisher Scientific (Hampton, N H , USA). A l l were of HPLC grade except for formic acid and water which were distilled in-house. 2.2 Methods 2.2.1 BAC-Crosslinked Gels Gel Casting/Electrophoresis Prior to the preparation of the acrylamide mixture, the gel-casting apparatus (Mini-Protean III, Biorad, Hercules, CA, USA) was assembled with the plates and incubated at 40-50°C for 10-15min. Heating the plates was an important preparation for casting dissolvable BAC-crosslinked gels [Hansen, 1981]. Previously, other groups have dissolved the crosslinker and monomer together and stored the mixture as a 20% stock solution [Hansen, 1976; Flyer et al, 1982]. Unfortunately, we have noticed that precipitation of the crosslinker occurs at room temperature over time and recommend against this. Instead, B A C was dissolved in 30% acrylamide at 50-60°C just prior to polymerization at a ratio of 37.5:1 (monomerxrosslinker). This ratio was selected since it is the most commonly used ratio for standard acrylamide/bis-acrylamide gel mixtures. For 5mL of a typical 10% mixture, 13mg of B A C was dissolved in 1.63mL 30% 27 acrylamide. The elevated temperature was necessary for B A C solubility and took 5-lOmin to complete. Once the crosslinker had dissolved, 1.5M Tris/HCl (pH 8.8) was added to a final concentration of 375mM. The mixture was then equilibrated to 45°C in a water bath prior to catalyst addition. T E M E D and APS were added to final concentrations of 2.5 and 0.02%, respectively, and the mixture was allowed to polymerize at 45°C for one hour. Before samples were run, the gels were pre-electrophoresed to remove contaminants such as unpolymerized acrylamide and excess TEMED since these are known to alter proteins during electrophoresis [Garzotti et al, 1998; Chiari et al., 1992]. Gels were pre-electrophoresed in 375mM Tris/HCl (pH 8.8) at 100V and 50mA per gel for one hour to remove contaminants. This buffer was then discarded, and the tops of the gels were rinsed with water prior to casting of the stacking gel. The stacking gel consisting of 5% bis-acrylamide/acrylamide, 125mM Tris/HCl (pH 6.8), 0.1% SDS, 0.1% APS, and 0.1% TEMED was then cast and samples were run without thiol-reducing agents at 125-200V and 30mA per gel in standard Tris/Glycine/SDS running buffer. Protein Isolation After electrophoresis, bands were stained with Coomassie R-250 and destained as previously described [Sambrook et al, 1989]. Bands were then excised and cut into small pieces (ca. 1mm x 1mm) using a scalpel (Fig. 2.1). Bands were dissolved in solubilization buffer containing lOmM Tris/HCl (pH 8.1) / 1% p-mercaptoethanol / 0.1% SDS at a ratio of 5mL buffer per gram of gel at 60°C for 30min with vigorous shaking. 28 Figure 2.1 Protein Isolation from a Dissolvable Gel. Workflow showing the general procedure for protein isolation and analysis using BAC-cross-linked SDS-PAGE gels. Cut out band(s) of interest BAC-crosslinked SDS-PAGE gel DTT reduces protein disulfide bonds and dissolves gel matrix > Exact mass of protein isoform Database search + Identification m/z Old Protein purification In-solution ^ Digestion m/z Peptide Mass Fingerprint 1. Enzymatic processing (dephosphorylation) 2. Digestion with other enzymes 3. Greater recovery of certain peptides 29 Since the reducing agent was the reductant for gel dissolution, 0- mercaptoethanol may be substituted with DTT. The pH of the band was an important consideration during dissolving since the optimal pH of disulfide exchange is around 8.3 [Hansen, 1980]. Even the most soluble bands which had been properly polymerized appeared completely insoluble in the presence of thiol-reducing agents at an acidic pH. Thus maintaining a slightly basic pH was a requirement for band solubility. A stronger buffer than the one recommended may be substituted since it will be removed in subsequent steps. Once dissolved, the resulting solution was very viscous due to the long chains of polyacrylamide, which became the major contaminant. The bulk of the polyacrylamide contamination was removed with a mixture of 1% l,l,l,3,3,3-hexafluoro-2-propanol (HFIP) in isopropanol in a ratio of 3 parts HFIP/isopropanol to 2 parts dissolved band solution. The mixture was then vortexed for lOsec and centrifuged at 10 000 rpm for lOsec to pellet the precipitated polyacrylamide material. The supernatant was then loaded into a Microcon spin column (Millipore, Billerica, M A , USA) with a 10 000 molecular weight cut-off filter and centrifuged as per manufacturer's instructions. The spin column was washed twice with 0.025% SDS to remove organic solvents, P-mercaptoethanol, salts, and trace polyacrylamide. The protein-containing supernatant was precipitated with 4 parts chilled acetone at -80°C for 60min. Afterwards, centrifugation was done at 14 OOOrpm at 4°C for 25min to pellet the protein. The supernatant was discarded and the pellet washed with a solution of 80% acetone / 20% methanol + water + acetic acid (50/47/3) and centrifuged again [le Maire et al, 1993]. The pellet was re-centrifuged at 14 OOOrpm for 20min and the supernatant 30 was discarded. The sample was then re-dissolved in an appropriate solution dependent on the type of intended analysis or experiment. Enzymatic Processing Dephosphorylation experiments were done using Calf Intestinal Alkaline Phosphatase (CIP) which was known to remove phosphate groups from serine, threonine and tyrosine residues in proteins and 5' phosphates from nucleic acids and nucleotides. Approximately 20units' of enzyme were used to dephosphorylate 20 ug of phosphoprotein. Samples were dissolved in the supplied NEBuffer3 (New England BioLabs) and incubated at 37°C with CIP for one hour. The reaction was quenched by adding formic acid to a final concentration of 1%. Samples were then acetone precipitated with 4 parts chilled acetone and incubated at -80°C for 60min. The protein was then pelleted by spinning at 14 OOOrpm and 4°C for 20min. The supernatant was discarded and the protein redissolved in 50% acetonitrile / 49% water / 1% formic acid for intact mass measurement via electrospray ionization with a Q-TOF. In-Solution Digestion Proteins containing disulfide bonds were reduced and alkylated prior to digestion for full denaturation and to prevent peptides from being linked together. First, samples were dissolved in the appropriate digestion buffer - either 50mM ammonium bicarbonate or lOmM Tris/HCl (pH 8.1). With the protein concentration between 0.5 - l u M , the sample Was heated at 90°C for 20min. Following the heating, the sample was incubated 1 one unit is defined as the amount of enzyme which hydrolyzes lumol of p-nitrophenylphosphate to p-nitrophenol in a total reaction volume of lmL in lmin at 37°C in 1M diethanolamine-HCl (pH 9.0) with 0.5mM MgCl 2 and lOmM p-nitrophenylphosphate. .31 in an ice/water mixture to quench the denaturation process. Next, the disulfide bonds were reduced by addition of DTT to a final concentration of ImM. The sample was again heated at 56°C for 60min to complete the reaction. Once cooled to room temperature, iodoacetamide was added to a final concentration of ImM and allowed to react in darkness for 45min at RT. Sequencing Grade Modified Trypsin and endoproteinase Glu-C were added at 30:1 and 50:1 (substrate:enzyme) ratios, respectively, and allowed to digest overnight at 37°C. Because chymotrypsin was much less specific than the previous two enzymes, it was used at a 100:1 ratio and only allowed to digest for 4 hours at 25°C. For this digestion, quenching with 5% formic acid (final concentration) was necessary to prevent over-digestion. In-Gel Digestion After electrophoresis and staining, bands of interest were excised from a BIS-crosslinked gel and cut into small pieces (ca: 1mm x 1mm) using a clean scalpel. The pieces were transferred to a microcentrifuge tube and washed with 100-150uL water for 5min at room temperature. The particles were spun down and the liquid was removed. Acetonitrile equal to approximately 3-4 times the volume of the gel pieces was added and incubated for 10-15min until the gel pieces shrank, became white and clumped together. The gel pieces were spun down again to remove the liquid and dried completely in vacuum centrifuge at 40°C for 5-10min. A solution of lOmM DTT/lOmM Tris/HCl (pH 8.1) was used to swell the pieces with enough liquid to cover them. Reduction took place for 30min at 56°C to reduce the 32 protein and was done even though proteins were reduced prior to electrophoresis. The pieces were spun down again to remove the liquid and shrunk with acetonitrile to remove excess DTT from the gel. Normally, gel pieces were dried completely in a vacuum centrifuge following the acetonitrile step, however this was skipped after reduction to prevent re-oxidation of disulfide bonds. After shrinking, the acetonitrile was replaced with a solution of 55mM iodoacetamide/10mM Tris/HCl (pH 8.1) and incubated at room temperature for 20min in complete darkness to alkylate free thiols and prevent disulfide bond reformation. The iodoacetamide solution was removed and the gel pieces were washed with 150-200uL lOmM Tris/HCl (pH 8.1) buffer for 15min. The gels were spun down and the liquid removed. The pieces were shrunk with acetonitrile and dried completely in a vacuum centrifuge. The gel particles were re-swollen in digestion buffer containing 1 OmM Tris/HCl (pH 8.1), 5mM CaCb, and 12.5ng/uL trypsin at 4°C for 30-45min. Any remaining liquid not absorbed by the gel was removed to prevent excess trypsin autolysis. The same buffer formulation lacking trypsin was added to cover the gel pieces and keep them hydrated during enzymatic cleavage. The samples were allowed to digest overnight at 37°C. For in-gel digestions involving endoproteinase Glu-C or chymotrypsin instead of trypsin, all steps were identical except that the chymotryptic digestion only took place for 4 hours before it was quenched with 1% formic acid. After digestion was complete or quenched, the peptides were extracted to maximize the yield. First, the tubes were spun down to collect any liquid in the cap of the tube and an additional 10-15uL of lOmM Tris/HCl (pH 8.1) was added. This solution was incubated at 37°C for 15min. Next 50uL of acetonitrile was added to shrink the 33 pieces at, again, 37°C for 15min. The samples were spun down briefly and the peptide-containing supernatants were transferred to separate tubes. For the final extraction step, 50uL of 5% formic acid was added to the particles and incubated at 37°C for 15min. Twice this volume of acetonitrile was then added and the final shrinking of the gel pieces took place at 37°C for 15min. The samples were again spun down and the supernatants were combined. The supernatants were either dried completely in a vacuum centrifuge or frozen and lyophilized prior to MS analysis. Coomassie Staining After electrophoresis, gels were submerged in a solution of 0.1% Coomassie R-250 / 10% acetic acid / 40% methanol for 2 hours at room temperature or were heated in a microwave for 30sec and incubated for 30min at room temperature for faster staining. The stain was removed and replaced with 10% acetic acid / 40% methanol solution for destaining. The destaining solution was changed repeatedly with fresh solution over several hours until the background of the gel became clear. The gels were stored in destain until bands were excised for analysis or purification [Sambrook et al, 1989]. Mass Spectrometry Intact proteins were analyzed using either the in-house MALDI-TOF/TOF (4700 Proteomics Analyzer, Applied Biosystems/MDS Sciex, Concord, ON, Canada) or QStar X L (Applied Biosystems/MDS Sciex) instruments. For MALDI-TOF analysis, protein samples were dissolved in 1% formic acid and mixed 1:1 with a saturated solution of sinapinic acid (3,5-dimethoxy-4-hydroxy-cinnamic acid, Sigma-Aldrich) matrix in 50% 34 acetonitrile / 5% formic acid. Next, 0.7uL of this was spotted onto a clean target and allowed to dry on the bench. Carbonic Anhydrase (Sigma-Aldrich) was also spotted in the same fashion and used as an external standard for calibration purposes. The samples were run in positive linear mode with 2000 shots summed per spectrum. For protein mass measurement on the Q-Star, isolated proteins were dissolved in 50% acetonitrile / 49% water / 1% formic acid solution. Samples were loaded into individual nano-electrospray needles coated with gold and palladium and mounted onto a commercial nano-spray interface (MDS Sciex). The instrument was controlled with the supplied Analyst QS vl.O software (Applied Biosystems). Ion spray voltages varied from 1000 - 1200V in TOF-MS mode with declustering and focusing potentials of 50 and 250V, respectively. Each measurement was summed with 5min acquisition times and deconvoluted using BioAnalyst v l . l software (Applied Biosystems/MDS Sciex) which was integrated into the control software. Calibration of the TOF was done with polypropylene glycol standard. For MS analysis of enzymatic digests, samples were prepared by mixing 1:1 with a saturated solution of a-cyano-4-hydroxycinnamic acid in 50% acetonitrile / 45% water / 5% formic and spotting 0.7uL of this mixture onto a clean target. The spots were allowed to dry at ambient temperature and then loaded into the spectrometer. The MALDI-TOF was used in positive reflectron mode with 2000 shots summed per spectrum. Bradykinin fragment 1-9 (Sigma-Aldrich) and A C T H fragment 18-39 (Sigma-Aldrich) had masses of 1059.5Th and 2464.2Th, respectively, and were used as external calibrants for all peptide-containing samples. 35 2.2.2 QTrap Analysis of H S A Chemical Acetylation HSA was first reduced and alkylated for complete denaturation prior to acetylation of lysine side chains with acetic anhydride. For reduction, 200u.g HSA was dissolved to a final concentration of 1.5uM in 2mM DTT / 20mM sodium borate (pH 9.0). The solution was heated to 60°C and kept at constant temperature for 60min. Once the sample had cooled back to room temperature, iodoacetamide was added to 2mM final concentration and allowed to alkylate for 45min at RT in darkness. Next, to further increase the pH above 10, sodium hydroxide was added to 300mM final concentration followed by l.OuL acetic anhydride (~50mM final cone). Since the reaction was exothermic, acetylation took place in a cold room at 4°C or, alternatively, in an ice/water mixture for 30min [Fazili et al, 1993]. The reaction was quenched by addition of formic acid to a 5% final concentration. Acetone precipitation was used for sample clean-up. Briefly, 1200uL chilled acetone was added and incubated at -80°C for 60min. Afterwards, the protein was centrifuged at 14 OOOrpm at 4°C for 20min. The acetone was carefully removed and the pellet was dried under a stream of nitrogen to remove excess acetone. The protein pellet was dissolved in 8M urea by vigorous vortexing. Liquid Chromatography LC-MS/MS experiments were carried out on an L C Packings Ultimate nano-LC system coupled to a Famos autosampler (Sunnyvale, CA, USA). The system loaded the sample on a C18 pre-column for sample clean-up for 5min, then switched to in-line where the sample was eluted off the pre-column onto a Dionex C-18 Pepmap 75 um x 36 150mm analytical column (Sunnyvale, CA, USA) for separation. The mobile phases were 95:5 water: acetonitrile with 0.1% formic acid (A) and 20:80 water:acetonitrile with 0.1%> formic acid (B). One hour gradients ramped mobile phase make-up from 95:5 to 30:70 (A:B) for elution of peptides bound to the C18 column. The eluant was electrosprayed from a New Objectives Picotip emitter (Woburn, M A , USA) with a lOum orifice by application of a 2500V ion spray voltage directly into the QTrap. Mass Spectrometry A l l mass spectrometry experiments were carried out on a QTrap (Applied Biosystems/MDS Sciex) hybrid triple quadrupole/linear ion trap mass spectrometer, using Analyst 1.4.1 control software. Conditions held constant between experiments included the 2.5kV ion spray voltage, curtain gas setting of 20, collision gas setting of 8, and 30V declustering potential. More specifically, all Enhanced Multiply Charged (EMC) and Enhanced Mass Spectrometry (EMS) scans were set to 10V collision energy, Q3 entry barrier of 8V and were scanned from 500-1500Th. A l l Precursor Ion (PREC) scans were set to 70V collision energy, 0.25amu step sizes, and were also scanned from 500-1500Th. To detect peptides containing the Ns-acetyl-lysine modification, the PREC scan was locked on the precursor of 126.1amu. Selection of peptides was controlled by Information Dependent Acquisition (IDA) which chose the three most intense signals and was set for a minimum threshold of 5000 counts that excluded isotopes within a 4amu window and former target ions for 30sec to prevent repeated fragmentation of very intense peptides. Enhanced Resolution (ER) scans had step sizes of 0.03amu, trap fill times of 50ms, Q0 trapping on, Q3 entry barrier of 8V, collision energies of 10V, and 37 scan rates of 250amu/s. Finally, Enhanced Product Ion (EPI) scans were done at 0.12amu step sizes, 4000amu/s scan speeds, QO trapping on, Q3 entry barrier of 8V, collision energies of 70V, 50ms trap fill times and scanned from 50-1500Th. Experimental Variables Various settings within individual scan types were altered in certain cases for EMC, EMS, and PREC scans. For the E M C and EMS scans, the following settings were varied: Q3 scan speed from 4000-250amu/s, QO trapping, and linear ion trap fill times from 50-0.01ms. A n option available with EMS scan but not the E M C was Dynamic Fil l Time (DFT) which was toggled at 50, 5 and 0.01ms trap fill times. As for PREC scan settings, varied parameters included the dwell time which was changed from 0.625 up to 2.5ms, and resolutions across Q l + Q3 were varied from low to unit. During testing of PREC scan sensitivity, the sample concentration was also varied by diluting from lOOOnM full concentration of a 1:1 (acetylated HSA tryptic digest: Markl2 standard tryptic digest) mixture down to 250nM with 0.1% formic acid. 38 CHAPTER 3: MASS SPECTROMETRY ANALYSIS OF INACT PROTEINS ISOLATED FROM DISSOLAVBLE SDS-PAGE GELS 3.1 Rationale and Hypothesis Post-translational modifications (PTMs) found on proteins hold one of the keys to understanding the regulation of their biological function including that of many signal transduction pathways. These additions may affect a protein's stability, distribution, structural conformation, physical and chemical properties, and activity which, in turn, determines their cellular function. Examples of the biological consequences of modifications include: phosphorylation for signal transduction, fatty acid additions for membrane anchoring, ubiquitination for proteolysis, and glycosylation for cell-cell interactions. Although the importance of PTMs of cannot be disputed, their identification and characterization on a proteomic scale has been held back due to the lack of a dependable, all-purpose method for their analysis. For protein identification from SDS-PAGE bands, in-gel digestion is the method of choice in proteomic studies. The process relies on identification based on peptides -either by peptide patterns in the PMF approach or by L C - M S / M S sequencing of individual peptides. While it is possible to identify and localize modifications using these approaches, obtaining the modified peptide(s) from a protein using the in-gel digestion procedure is never guaranteed. Peptides are obtained largely by chance since they must diffuse passively from the gel band and so the only way of ensuring a modified peptide is obtained is to isolate the entire protein from the gel and digesting it in solution. With this approach, variability due to diffusion from the gel band is eliminated since the intact 39 protein is isolated. Furthermore, isolation of intact protein from gel bands has several other attractive possibilities including: exact mass measurements of the entire protein isoform, enzymatic processing for P T M scouting, and increased sequence coverage and recovery of peptides from in-solution digestions. In this study, the dissolvable polyacrylamide cross-linker, bis-acrylylcystamine (BAC) was used in an SDS-PAGE gel for protein electrophoresis and subsequent protein purification and mass spectrometric analysis. As a post-translationally modified model protein, a-casein was used as it possesses numerous possible phosphorylations on serine residues. With this in mind, we hypothesized that we would be able to obtain a very precise, intact mass of a-casein using this dissolvable SDS-PAGE system. Moreover, it was also postulated that, using a phosphatase to remove the phosphate groups previously reported on a-casein, we would be able to deduce its modification status from the mass difference of the unprocessed form. In addition, we hypothesized that since intact protein would be isolated from SDS-PAGE bands, we would be able to digest with a greater variety of enzymes and that the digests would contain greater amounts of peptides compared to traditional in-gel digestions. 3.2 Results and Discussion 3.2.1 T E M E D / B A C Relationship Creating gels which dissolve in the presence of thiol-reducing agents requires special care and attention to certain conditions. The polymerization kinetics must be 40 drastically altered due to a competing side-reaction which yields insoluble gels. Spurious thiol crosslinks, consisting of C-S bonds, appear from a reaction of the acrylamide radical and the disulfide bond within the B A C molecule itself [Hansen, 1980]. The side reaction yields a new version of the crosslinker which is able to crosslink two chains of polyacrylamide together but is, unfortunately, stable in the presence of p-ME/DTT. Very rapid polymerization with increased temperature and concentration of catalyst are necessary requirements for successful B A C polymerization. Gels typically solidify in less than one minute vs. 10-20min for conventional gels, thus immediate pouring is critical. The first challenge in this study was to determine a good gel recipe which produced gels of similar quality to standard BIS-crosslinked gels and optimal conditions for polymerization which produced soluble gels with few spurious crosslinks. We determined that a key relationship existed between T E M E D and B A C crosslinker concentrations which, largely, governs both gel quality and solubility. TEMED counteracts the negative side reaction responsible for spurious crosslinks and insoluble gels, possibly by simply increasing the rate of polymerization. Thus, the more B A C crosslinker is used, the more T E M E D is needed to maintain subsequent gel solubility. There is, however, an upper limit to the amount of T E M E D which may be used for certain percentage gels as T E M E D is a chain initiator for acrylamide. The more TEMED is used, the shorter polyacrylamide chains become, which may compromise gel quality to where gelation may not appear to take place at all [Flory, 1953]. Conversely, i f too little TEMED is used, the competing side reaction proceeds to create enough spurious cross-links to create a gel which does not readily dissolve. Using less T E M E D and too little 41 cross-linker also compromises gel quality because gels become too weak and fragile to handle due to insufficient cross-linkage. Thus, there is a fine balance between the amount of T E M E D needed in relation to the amount of B A C used. Since there is a practical limit to the amount of TEMED used, this limits the amount of B A C which may be used as well. Previous studies have shown that gel concentrations of >12% do not readily dissolve due to the amount of B A C required [Anker, 1970]. It is the high TEMED concentrations required which make these gels impractical for use. In Figure 3.1, optimal T E M E D / B A C concentrations are shown for a 10% gel. The box outlines the concentrations which give gels of good quality and excellent solubility when reducing agents are added. Using too little TEMED and/or too much B A C causes gels to become insoluble due to the competing side reaction. Conversely, too much T E M E D and/or too little B A C creates gels of poor quality which tend to break easily or fail to fully polymerize. 3.2.2. Pre-Electrophoresis Due to high amounts of TEMED necessary to create soluble gels, pre-electrophoresis of the separating gel only in its own buffer (ie. 375mM Tris/HCl, pH 8.8) is required to remove charged impurities left over from polymerization [Seymour et al., 1989]. We have noticed that proteins run in gels which have not undergone pre-electrophoresis have mass shifts of several hundred Daltons indicating possible extensive modification during electrophoresis. The spectra in Figure 3.2 are of a-casein analyzed on the MALDI-TOF in linear mode. As a control, standard a-casein was run (Fig. 3.2A) 42 Figure 3.1 Balancing BAC and TEMED Concentrations. Trendline showing optimal concentrations of T E M E D and B A C crosslinker which yield gels of good solubility and quality for 10% acrylamide mixtures. The box outlining the trendline shows the region where optimal concentrations are found. 0.4 H 0 0.2 0.4 0.6 0.8 1.0 % BAC (w/v) 43 and compared to B A C gel-extracted protein from gel run with (Fig. 3.2B) and without (Fig. 3.2C) pre-electrophoresis The singly charged versions of the standard and the sample extracted from the gel which had undergone pre-electrophoresis are in good agreement with each other at 23 604Th and 23 602Th, respectively, indicating that no modification occurred during the course of electrophoresis and throughout the remainder of the purification. The sample isolated from the gel Which did not undergo this step shows a mass shift of nearly lOOODa, suggesting that extra modifications were accumulated at some point(s) during the experiment. Moreover, the peak width of this sample is much broader and lacks sharpness at its apex which is present in the others. This increased peak width and decreased resolution is indicative of much heterogeneity in the sample originating from electrophoretic adducts. Furthermore, when the non-pre-electrophoresed sample was run on a Q-TOF instrument equipped with an ESI source, there was no defined distribution of charge states as anticipated and thus no exact intact mass could be calculated. Instead, the data showed low-intensity broad rolling peaks which also supported the notion of widespread heterogeneity within the sample. Protein adducts from SDS-PAGE have been well documented and studied by mass spectrometry and N M R to determine the exact nature of the modification. Most notably, the reaction during electrophoresis which showed consistent adduction was that between cysteine and unpolymerized acrylamide monomer to form cysteinyl-S-/?-propionamide. In fact, it is known that polymerization rarely exceeds over 90%, leaving a considerable amount of reactive acrylamide available for adduct formation. This may be especially relevant to the polymerization of B A C cross-linked gels as the gelation is very fast compared to traditional gels. Unpolymerized acrylamide monomers separated 44 Figure 3.2 Importance of Pre-electrophoresis. MALDI-TOF MS of (A) standard, (B) gel-extracted with pre-electrophoresis and (C) gel-extracted a-casein without pre-electrophoresis. 45 from each other by the polymerized gel matrix may not react as efficiently to produce chains because of physical constraints. Since BIS-crosslinked gel mixtures polymerize slower, acrylamide monomers may have the opportunity to react more completely. Interestingly though, a-casein does not contain any cysteine residues other than one present in the signal sequence which should be removed in the commercially obtained protein. Since the reaction of these adducts proceeds via nucleophilic attack of the free thiolate of Cys with the double bond of the acrylamide monomer, it is conceivable that other nucleophilic groups can participate in the absence of thiolate ions. Terminal primary amines and lysine and histidine side chains may also be candidates for possible nucleophiles capable of reacting with acrylamide. 3.2.3 Resolution Comparison Trying to match all the specifications of the traditional SDS-PAGE gel, we compared the resolutions between BIS and BAC-crosslinked gels with standard proteins. BIS and BAC-crosslinked gel were cast using identical acrylamide:monomer ratios (37.5:1) and run with a set of standard proteins under non-reducing conditions (Figure 3.3A). Equal loadings of proteins from a variety of molecular weights were run to show that the resolutions of both sets of standards are roughly equal. The bands in the B A C gel seem to be less defined than the BIS cross-linked gel, particularly the middle ovalbumin band. This may be due to the fact that the cross-linker itself is longer than bis-acrylamide which creates a slightly larger pore size. This slightly larger pore size may be responsible for the decreased band sharpness. Another possibility is the fact that B A C itself is quite hydrophobic, owing to the bulky disulfide group [Gelfi et al, 1992]. Thus, a possible 46 Figure 3.3 Resolution Comparison Between BIS and B A C Gels. (A) Non-reduced standard markers run on BIS- and BAC-crosslinked SDS-PAGE gels of 37.5:1 (monomer: crosslinker) ratios for a comparison of resolution. (B) SDS-PAGE gel of non-reduced standards further showing the resolution capabilities with proteins of similar molecular weights. A kDa Protein 66 Albumin, Bovine Serum 45 Ovalbumin 29 Carbonic Anhydrase BIS BAC 37.5:1 37.5:1 B kDa 47.2 23.6 14.3 12.4 Protein a-casein (dimer) a-casein (monomer) lysozyme cytochrome C BAC 37.5:1 47 hydrophobic effect between the matrix and hydrophobic side chains on the protein during electrophoresis may result in an interaction which contributes to the observed band fiizziness. To further demonstrate the resolving capabilities, a different set of proteins run under non-reduced conditions is shown in Figure 3.3B. As shown, the lysozyme and cytochrome C bands at 14.3 and 12.4 kDa, respectively, are clearly resolved using a 10% BAC-crosslinked gel. 3.2.4 Electrophoresis with Reducing Conditions Traditionally, SDS-PAGE gels are run with P-mercaptoethanol or dithiothreitol (DTT) in the sample buffer to completely denature the analyte and allow it to adopt a random coil conformation. This, and the fact that SDS binds most proteins in largely the same ratio, ensures that proteins are separated based on their molecular weight rather than on their tertiary structure or number of charges. Regrettably, reducing conditions must be completely avoided for successful electrophoresis using BAC-cross-linked gels. Reducing agents cause the gel matrix to dissolve prematurely and seriously compromise band quality, resulting in smearing and poor band sharpness. In Figure 3.4, a-casein was run on a 10% B A C gel with and without various concentrations of DTT to test how much reducing agent could be tolerated. With no DTT present in the first two lanes as controls, the bands appear sharp and defined, as expected. However, with as little as 5mM DTT, a shift in mobility and band smearing beginning from the very top of the lane is evident. The shift in mobility is expected since the proteins are now completely unfolded compared to the non-reduced lanes where they are more structurally compact and can 48 Figure 3.4 Reducing SDS-PAGE with BAC Gels. A BAC-cross-linked gel run with equal amounts of a-casein and increasing concentrations of thiol-reducing agent, dithiothreitol (DTT) was run. Reducing agents commonly included in SDS-PAGE sample buffer such as P-mercaptoethanol or DTT must be avoided during B A C gel electrophoresis. mM DTT 0 0 5 11 22 44 88 175 350 700 49 travel further. While it is true that reduced proteins may bind more SDS molecules which would contribute to more overall negative charge, the complete unfolding and linearity of the protein seems to prevent migration which is faster than the non-reduced sample. As the concentration of reducing agent is increased stepwise, band distortion becomes more pronounced until complete destruction of the gel matrix takes over around 1 OOmM and adjacent lanes melt together. It has been previously reported that thiol-reducing agents may be used during electrophoresis of BAC-cross-linked separating gels so long as the stacking gel remains BIS-cross-linked [Flyer et al, 1992]. The idea being that the stable BIS stacking gel acts as a physical barrier and so reducing agents would not run with the sample into the separating gel. The figure shown here had incorporated the BIS-cross-linked stacking gel and yet shows evident gel destruction using DTT during electrophoresis. Presumably, the pH of running buffer was high enough for thiol deprotonation and caused DTT to migrate through the gel, destroying the matrix in the process. To run completely reduced proteins on B A C gels, the sample would first need to be reduced and alkylated followed by purification (ie. acetone precipitation, C4 chromatography, etc.) to remove the reducing agent prior to electrophoresis. 3.2.5 Precipitation of Dissolved Polyacrylamide with Organic Solvents Once the sample was run on a properly polymerized BAC-crosslinked gel, stained bands were excised and dissolved using a buffered solution containing a thiol reducing agent such as DTT or P-mercaptoethanol. Once dissolved, the resulting solution was rather viscous due to the long chains of polyacrylamide. This viscosity prevented any 50 sort of column chromatography, such as HPLC, because of the unacceptably high back pressure. One possible way of alleviating this would be to add more solubilization buffer at the expense of further diluting the protein. Instead of using this approach, we noticed from other experiments that the addition of organic solvents to a dissolved polyacrylamide solution caused the precipitation of polyacrylamide. To test whether or proteins would co-precipitate, BAC-crosslinked bands containing a-casein were tested. After dissolving the bands in a buffered solution of 1% P-mercaptoethanol, either acetonitrile or isopropanol was added to cause precipitation of the dissolved polyacrylamide. The supernatants were then loaded onto a regular 10% BIS-acrylamide gel and stained with Coomassie to assess protein recovery. As a control, two dissolved bands were directly loaded in the first two lanes without undergoing any organic solvent treatment. In an attempt to improve protein solubility, hexafluoroisopropanol (HFIP) was added since it is known to be a potent solubilizer of proteins and prevents precipitation at higher concentrations of organic solvent [Higham et al, 2000]. As shown in Figure 3.5, no protein was recovered after precipitation of the polyacrylamide with acetonitrile even in the presence of HFIP, indicating that a-casein was co-precipitated. In contrast, a-casein was soluble in isopropanol and its recovery was shown to be enhanced with the addition of HFIP (Fig. 3.5). Both solvents removed polyacrylamide with approximately equal efficiency; however isopropanol caused significantly less sample loss and was therefore used for subsequent experiments. 51 Figure 3.5 Precipitation of Dissolved Polyacrylamide and Protein Solubility. A 10% BIS-acrylamide gel showing the recovery of a-casein from BAC-crosslinked bands after precipitation with either acetonitrile or isopropanol for dissolved polyacrylamide removal. Hexafluoroisopropanol (HFIP) was also added to a final concentration of lOOmM. Dissolved B A C bands containing a-casein which did not have the dissolved polyacrylamide removed were loaded as controls in the first two lanes. + + + + HFIP Added J ^ J Control Bands Loaded Acetonitrile Isopropanol 52 3.2.6 Extracted Proteins Once protein is recovered from a BAC-cross-linked gel, its mass can be accurately measured, an experiment which is not possible with standard gels. Additional experiments such as enzymatic processing, and in-solution digestions with enzymes other than trypsin for enhanced recovery of peptides become possible. Several standard proteins of varying molecular weights such as myoglobin (~17 kDa), a-casein (~24 kDa), and fetuin (-47 kDa) were run on B A C gels, excised, purified and analyzed on either a MALDI-TOF (Applied Biosystems/MDS Sciex) or a Q-TOF (Applied Biosystems/MDS Sciex). Myoglobin obtained from horse heart (Sigma-Aldrich) is a 153 amino acid, 16951 Da protein belonging to the globin family and, more specifically, the two-domain flavohemoprotein subfamily. Its primary purpose is as a reservoir for the supply of oxygen and also to facilitate oxygen movement within muscle. To make this possible, it contains two heme groups with iron molecules bound to histidines at positions 64 and 93 of its sequence. Myoglobin was successfully purified from a B A C cross-linked gel and analyzed by MALDI-TOF MS (Figure 3.6). The standard was run as a control (Fig. 3.6A) along with the gel-extracted sample (Fig. 3.6B) with prior calibration of the TOF. Singly (16943Th), doubly (8476Th), and triply (5652Th) charged forms are clearly evident in both samples. As shown, the standard (16943Th) agrees well with the predicted (16951Th) mass along with several peaks following the base peak. Another successfully purified protein was bovine fetuin (Sigma-Aldrich) which is a 359 amino acid, 38 419 Da secreted glycoprotein. Its biological features include promoting endocytosis as well as possessing some opsonic properties. Thus, it plays a 53 protective role in serum. The protein possesses several carbohydrate moieties O-linked to serine and threonine as well as N-linked to asparagine, which create heterogeneity and thus many isoforms. This heterogeneity is clearly shown in the MALDI-TOF MS spectra of standard (A) and gel-extracted (B) fetuin (Figure 3.7). It is evident that both spectra lack any sharp, clearly resolved peaks for the singly-charged and doubly-charged peaks at 47 kTh and 23.5 kTh, respectively. Instead, very wide peaks characteristic of many different isoforms of fetuin are present and give an idea of the overall complexity of the sample. We can infer that there must be many different locations and variations where sugar groups are attached to the protein and are thus responsible for creating many masses. Alternatively, since several N-linked sugars are present, there could also be one complex heterogeneous sugar. Having a distribution of masses present within the same sample creates broad peaks with little resolution instead of sharply defined peaks. Lastly, gel-extracted a-casein was purified as described in the Methods section and its exact mass measured on a Q-TOF instrument (QStar X L , Applied Biosystems/MDS Sciex). With 199 amino acids, it is a secreted bovine protein found in milk which functions to transport calcium phosphate. Structurally, it lacks any disulfide bonds, or cysteine residues for that matter, and can have up to 10 phosphorylated serine residues. These phosphorylations can be removed with an alkaline phosphatase, and the exact mass of the native and enzymatically-processed form of gel-extracted a-casein can be measured to deduce the exact number of phosphorylations. Electrospray mass spectra for standard a-casein (Fig 3.8A) and gel extracted a-casein (Fig 3.8B) were used to obtain the intact masses. Since the mass spectral data for both contain many multiply-charged forms of a-casein, a mathematical calculation was 54 Figure 3.6 Purification of Myoglobin. MALDI-TOF mass spectra of standard (A) and gel-extracted (B) myoglobin from horse heart. The singly-charged first isoform of myoglobin in the standard at 16 946Th is in good agreement with the gel-extracted first isoform at 16 949Th. 8 960 12 971 16 982 20 993 Mass (m/z) 8 960 12 971 16 982 20 993 Mass (m/z) 55 Figure 3.7 Purification of Fetuin. MALDI-TOF mass spectra of standard (A) and gel-extracted (B) bovine fetuin. The; singly-charged mass of fetuin (~ 47kDa) was difficult to measure due to the vast heterogeneity of the sample. The doubly-charged signal is shown at ~ 24kDa. 56 Figure 3.8 Intact Mass Measurements of Unprocessed and Processed a-casein. ESI mass spectra of standard (A) and gel-extfacted (B) intact a-casein. Magnified deconvoluted spectra for both are also shown in (C) and (D), respectively, with insets showing the entire spectra. Magnified deconvoluted spectra for CIP-treated a-casein are shown for standard (E) and gel-extracted (F) samples, along with the corresponding insets containing the entire deconvoluted mass spectra. 600 1000 m/z 1400 1800 2200 600 1000 m /z 1400 1800 2200 2.35e4 2.37C4 m / z 2.39e4 2.41e4 2.35e4 2.37e4 m / z 2.39e4 2.41e4 2.290e4 2.300e4 m/z 2.310e4 2.320e4 2.290e4 2.300e4 m / z 2.310e4 2.320e4 57 applied to deduce the intact, singly charged mass. This process of calculating the mass from several multiply charged species is known as Bayesian reconstruction or deconvolution. The mass obtained for BAC-extracted a-casein of 23 616.46 ± 0.75Th (average of five measurements) (Fig. 3.8D) is in good agreement with the control at 23 615.49 ± 0.35Da (Fig. 3.8C). Examinations of the deconvoluted spectra of both samples reveal a remarkably similar pattern of peaks following the base peak (Fig. 3.8C, D). Some of these peaks, with approximate mass differences of 23Da and 39Da, are likely the result of sodium and potassium adducts with +22Da and +38Da, respectively, from the M + H peak. The peaks following the base peak in the gel-extracted sample seem to be more intense and less defined as the standard, possibly due to trace impurities. However, the masses of both samples are approximately ITh apart with errors which overlap. One can safely assume that no non-specific modification had occurred on the gel-extracted sample during the course of the electrophoresis or the purification steps that followed. Since a-casein is modified with several phosphorylations, we chose calf intestinal alkaline phosphatase to remove these modifications and show that this mass shift could be detected with intact protein. As shown in Figure 3.8, treating standard (E) and gel-extracted (F) a-casein with this phosphatase shifted the masses several hundred Th lower than the untreated samples (3.8C & D). The zoom-in of the deconvoluted peaks showed that most peaks surrounding the base peaks were almost identical suggesting that, again, no non-specific modification altered the protein (Fig. 3.8E, F). The number of phosphorylation sites can be determined by comparing the mass before and after phosphatase treatment. Both masses obtained from CIP-treated control standard and gel-58 extracted protein differed by only 1.22Th with little error. Finding the difference between the untreated and CIP-treated samples showed a shift of 640.90Th for the standard and 640.65Th for the gel-extracted protein. Since the mass difference associated with a single dephosphorylation event is 79.97Da, we can conclude that 8 serine, threonine and/or tyrosine residues are modified in the untreated protein. Indeed, a-casein has been previously shown to have up to 10 serine phosphorylations possible [UniprotKB/Swiss-prot entry P02662]. Also based on the fact that the patterns before and after dephosphorylation are very similar, we can deduce that the 8 sites of phosphorylation are stoichiometric. 3.2.7 Digestions The in-gel digestion is the standard method of protein identification in most proteomics laboratories. This method has inherent shortcomings which include limited recovery and/or varied digestion efficiency of certain enzymes compared to the in-solution digestion. To prove this, we compared the peptide recovery of a-casein after a tryptic digestion with in-solution and in-gel samples, a-casein was run on a 10% BIS-acrylamide gel stained with Coomassie and excised for in-gel digestion. The same protein was also run on a 10% BAC-crosslinked gel, extracted and purified for in-solution digestion. These were both compared to a standard a-casein in-solution digest as a control. The results of the tryptic digestions are shown in Figure 3.9 after analyzing all the samples on a MALDI-TOF-MS. The most intense signals in both in-solution samples are clearly present for standard and gel-extracted a-casein. For in-gel digestion, not only are the signals less intense, but two of the most intense peaks found in both in-solution 59 Figure 3.9 Gel-Extracted and In-Gel Tryptic Digestions. Tryptic digestions of standard (A), gel-extracted (B), and in-gel (C) a-casein. Peptides which are present in both standard in-solution and gel-extracted in-solution digestions, but absent in the in-gel digestion are indicated (*). 100n 90 80 '</> 60 aj 50 J 40 ^ 30 20 10 0 1267.69 1759.94 A * 2316.14 1384.71 * 1959.00 ,., i , 1,,, I. 4013.0 4013.0 1441.8 2084.6 m/z 2727.4 3370.2 4013.0 60 digestions appear to be completely absent and are indicated by a star (Fig. 3.9). Whether this is due to decreased digestion efficiency in gel or because the larger or more hydrophobic peptides diffuse less readily from the gel is difficult to assess. Moreover, there are fewer signals of low intensity in the in-gel sample compared to both in-solution samples. A complete list of all peptides obtained for all three tryptic digestions is shown at the top of Table 1. Several peptides of low intensity such as the ones found at 910.5Th, 1580.8Th, and 1871.9Th are not found in the in-gel samples but are present in both in-solution digestions. Obtaining increased sequence coverage is done by digesting with different enzymes to generate different, hopefully overlapping, peptide sets. Endoproteinase Glu-C, which cleaves on the C-terminal side of glutamic acid and, depending on the buffer, aspartic acid, and chymotrypsin, which cleaves on the C-terminal side of tryptophan, tyrosine, phenylalanine and methionine residues, are often used as substitutes for trypsin where arginine and lysine cleavage sites are scarce or potentially modified. In-gel digestions of proteins in SDS-PAGE bands are predominantly performed with trypsin compared to Glu-C and chymotrypsin, thus we tested whether digestion efficiency and the number of peptides obtained varied with the enzyme used. These results were compared to in-solution standard and in-solution BAC-extracted a-casein. Our findings are summarized in Table 1 along with the peptides obtained from the trypsin digestion. A comparison of the results in the in-solution standard column to in-solution BAC-extracted column shows that many of the peptides are present in both at roughly equal intensities. There are, however, some peptides with weak intensities near the baseline which are only present in the in-solution standard Glu-C and chymotrypsin digestions. A l l major signals 61 Table 1 Comparison of detected peptides from trypsin, Glu-C, chymotrypsin digestions of standard, BAC-extracted, and in-gel a-casein. Enzyme Peptide Mass (Da) Sequence Position # Missed Cleavages " In-Solution Standard In-Solution BAC-extracted In-Gel Digestion Trypsin 910.5 140- 147 0 + + — 946.5 5 0 - 57 1 - ++ — 1267.7 106- 115 0 +++ +++ +++ 1337.7 9 5 - 105 1 ++ + + 1384.7 3 8 - 49 0 +++ +++ +++ 1580.8 121 - 134 0 + + -1660.84 121- 134 0 ++ ++ + 1759.9 2 3 - 37 0 +++ +++ +++ 1871.9 119- 134 1 + + -1959.1* 119 — 134 1 +++ ++ + 2316.1 148- 166 0 +++ +++ -2332.1 c 148- 166 0 - - ++ Glu-C 876.5 208- 214 0 + + 900.5 134- 140 0 + -902.4 157- 163 0 + + -910.5 9 3 - 99 1 + - -913.5 105- 111 0 + + -1049.6 4 6 - 54 1 + — — 1222.7 c 205- 214 1 + — — 1440.7 100- 111 1 +++ ++ ++ 1449.8 3 4 - 45 0 +++ +++ +++ 1664.9 1 6 - 29 0 ++ — -1756.1 112- 125 0 +++ +++ +++ 1778.9 141 — 156 0 ++ ++ ++ 2120.2 1 6 - 23 1 +++ +++ +++ 2332.2 9 3 - 111 2 + + + Chymotrypsin 859.5 166- 171 2 + _ + 905.5 4 0 - 47 0 + + -948.5 107- 114 4 + -1052.5 3 9 - 47 1 + • -1099.5 161 — 168 1 + + — 1107.5 170- 179 1 + - — 1153.6 151- 160 2 ++ - -1217.7 107 - 116 5 +++ +++ — 1246.6 160- 168 3 . + - — 1262.6 161- 169 3 +++ +++ — 1270.6 169- 179 2 + — -1355.7 3 7 - 49 2 +++ +++ -1409.7 160- 169 4 ++ ++ — 1523.8 108- 119 5 + — — 1655.7 151 - 164 '3 + + — 1701.7 166- 179 3 +++ +++ -1864.8 166- 180 4 +++ +++ — 1877.0 1 6 - 31 0 + + + 2460.3 1 6 - 36 2 +++ +++ +++ 2763.6 1 6 - 38 3 ++ ++ — 2910.6 16- 39 4 +++ +++ -+++ = relative peak intensity > 25%; ++ - 25% > relative peak intensity > 5%; + = relative peak intensity < 5%; - = peak not present a. Missed cleavages based on the following enzyme specificities: Trypsin cuts C-terminal of K/R except before P; Endoproteinase Glu-C cuts C-terminal of E except before P; Chymotrypsin cuts C-terminal of F/Y/W/M/L except before P b. Known phosphorylated peptide c. Methionine S-oxidarion (MSO) 62 are present in the BAC-extracted samples, indicating that digestions from either are equivalent in the quantity of information provided. In-gel digestions, regardless of the enzyme used, consistently show fewer peptides and therefore provide less information for an equivalent amount of sample. As shown in Table 1, many of the major peaks found in trypsin and Glu-C digestions with in-solution samples were present in the respective in-gel samples. For tryptic digestions done in gel, 7 of the 11 peptides were detected compared to the samples done in solution. The intense tryptic peptide of 2316.1Th was completely absent in the in-gel digest but was found shifted to 2332. ITh with an oxidized methionine side chain. Other peptides found with low intensities for in-solution samples were lost using in-gel digestions, such as 910.5, 946.5, 1580.8 and 1801.9Th. As for Glu-C digestions, only 5 of the 14 peptides could be detected with the in-gel sample. Peptides of higher intensities were detected in all three digestions, however information was lost for low intensity signals with the in-gel sample. Low intensity peptides which could not be detected with the in-gel sample included peptides found at 876.5Th, 902.4Th, and 913.5Th. For chymotrypsin, very few peptides were detected for the in-gel samples with only 3 of the 21 total peptides available. Chymotrypsin was most efficient with samples which were in solution only. Many more missed cleavages are common for this enzyme considering the number of possible cleavage sites and the limited proteolysis times involved. Moreover, the digestion is done at room temperature for chymotrypsin compared to 37°C trypsin or Glu-C and also done at a much lower enzyme: substrate ratio. Digestions with trypsin, Glu-C and chymotrypsin were done at enzyme: substrate ratios of 25:1, 50:1 and 100:1, respectively. 63 3.2.8 Concluding Remarks In this chapter, a method for isolation of intact proteins from dissolvable SDS-PAGE gels was developed. Several proteins including a-casein (23.6kDa), fetuin (~47kDa) and myoglobin (17kDa) were run on the BAC-crosslinked gels, purified and analyzed by mass spectrometry. In particular, since a-casein is known to contain post-translational modifications, further analyses were done to study them using intact protein isolated from the dissolvable gels. Using a phosphatase to specifically remove phosphates and measuring the mass shift before and after such treatment, it was shown for a-casein that exactly 8 of the 10 possible sites of phosphorylation previously reported were modified. Further to being able to accurately measure intact protein masses and deduce the presence of post-translational modifications, overall digestion efficiency is also increased. Because digestion is now done in solution rather than using the standard in-gel approach, enzymes are no longer restricted by the gel matrix and peptides need not diffuse out of the gel either. This was shown by comparing in-gel with in-solution digested proteins which were either standards or B A C gel-extracted samples. A l l proteins digested in solution showed consistently higher yields of peptides when compared to the in-gel digested samples. 64 CHAPTER 4: SYSTEMATIC EVALUATION OF DATA ACQUISITION CRITERIA FOR COMPREHENSIVE ANALYSIS OF MODIFIED PROTEINS ON A HYBRID TRIPLE QUADRUPOLE/LINEAR ION TRAP INSTRUMENT 4.1 Rationale and Hypothesis In the previous chapter, proteins were isolated from a dissolvable SDS-PAGE gel system and, in the case of a-casein, the modification status was evaluated. While isolating intact protein from SDS-PAGE provides a means of physically ensuring modified proteins are extracted, determining the individual modification sites is the next challenge. Using acetylation of lysine side chains as a model, we tested the usefulness of a hybrid triple quadrupole/linear ion trap mass spectrometer to analyze both unmodified and post-translationally modified peptides within a single L C run. Triple quadrupole instruments are widely used for the detection and analysis of post-translational modifications using neutral loss and precursor ion scans for labile and stable modifications, respectively. While extremely robust and versatile, these instruments lack the sensitivity of, for instance, ion traps which are able to accumulate ions for a decreased limit of detection. The recently introduced hybrid triple quadrupole/linear ion trap (QTrap, Applied Biosystems/MDS Sciex) instrument combines all the functionalities of a standard triple quadrupole with the sensitivity found in ion traps by using the third quadrupole (Q3) as a linear ion trap (LIT). In addition to its increased sensitivity, several survey scans may be done sequentially within a single L C experiment to detect both modified and unmodified peptides for the comprehensive 65 analysis of post-translationally modified proteins. The most intense peptides are selected and subjected to fragmentation for sequence information. Using chemically modified human serum albumin as a model, we evaluated the usefulness of this instrument to sequence both unmodified peptides and peptides containing lysine acetylation using the combined survey scan approach in a single experiment. We were expecting all 10 modified peptides in the sample would be easily detected and sequenced by this instrument using this type of setup. Our preliminary findings indicated that 3 of the 10 modified peptides were not chosen for MS/MS and thus not sequenced when Enhanced Multiply Charged (EMC) and Precursor Ion (PREC) scans were used in combination within a single run. Thus even though we can increase the likelihood of obtaining modified peptides by, for example, isolating the intact protein from an SDS-PAGE gel, we are still not guaranteed that they will be sequenced. Attempting to use two survey scans sequentially to obtain information on all modified peptides as well as cover as many unmodified peptides as possible leads to difficulties in modified peptide sequencing. Possible explanations for missing modified peptides present in the sample during MS/MS selection are as follows: a. The PREC scan was not able to use trapping in Q3, it was not sensitive enough to detect or initiate MS/MS sequencing of all modified peptides. b. Combining both E M C and PREC scans together caused there to be a bias for selection of peptides from one scan over the other and so i f modified 66 peptides were not the most intense signals in the favoured scan, they would be missed. Thus, we surmised that maximizing the sensitivity of the PREC would be a possible solution to the problem of missed acetylated peptides. We hypothesized that by optimizing certain parameters within the PREC scan, such as QI + Q3 resolution, step size and dwell time at each step, it would become more sensitive. In addition, we speculated that i f there was a bias with respect to peptide selection from the E M C scan, it could be decreased or eliminated by detuning the scan. We hypothesized that by decreasing the time ions are trapped in Q3, the intensities of ions in the E M C should decrease proportionally. Combining both the optimized PREC scan with the detuned E M C scan should increase the number of modified peptides selected and sequenced. 4.2 Results and Discussion 4.2.1 Acetylation of Human Serum Albumin Human serum albumin (HSA) is the principal protein in blood plasma which serves mainly to regulate the colloidal osmotic pressure of blood. However, it also possesses the ability to bind small molecules such as water, calcium, sodium, potassium, fatty acids, and hormones. With an approximate weight of 67 500 Da, HSA contains seventeen disulfide bonds as well as N-linked glycosylations depending on the variant. It 67 also has 60 lysine residues, making it a good candidate for chemical acetylation as many possible modified peptides would result. We chose HSA as a model protein for this study over naturally modified proteins, such as histones, since it would provide several acetylated peptides after treatment with acetic anhydride and a relatively large amount would be available for the experiments. Acetylation of lysine side chains in proteins using acetic anhydride is a common procedure providing efficient modification. A vast excess of acetic anhydride must be used relative to the stoichiometric ratio of lysines in a given protein since the anhydride is prone to an aggressive hydrolysis side reaction in any aqueous solution. Acetic anhydride will cause the acetylation of not only the e-amino groups on lysine but also the a-amino groups on N-termini. We have noticed O-acetylation on serine, and threonine residues from using too much acetic anhydride. Such peptides are evident during MALDI-TOF analysis where numerous shifts of 42Da are present in addition to the original lysine - acetylated peptide. Since several over-acetylated forms of the same peptide may exist, this decreases the overall intensity of that peptide due to signal splitting. Thus, optimization of the acetic anhydride concentration used as well the reaction time was necessary to acetylate only the lysine side chains. HSA contains 60 lysine residues which were modified to different extents with acetic anhydride and then digested with trypsin to yield peptides for Q-Trap analysis. The varied extent of modification was intended as it more realistically depicted the modification status of many acetylated proteins in vivo [Huq et al, 2005]. Thus, the vast majority of lysines were not 100% acetylated, as can be clearly seen in Figure 4.1. Since trypsin can no longer cleave at a modified lysine due to the lack of positive charge, it was 68 Figure 4.1 Sequence of Chemically Acetylated Human Serum Albumin. Amino acid sequence of HSA showing detected acetylated peptides (I 1 ) obtained from acetic anhydride treatment followed by tryptic digestion along with their predicted molecular weights. Acetylated lysine residues are shown in bold italics. 10 20 30 1267 Da 40 50 60 MKWVTFISLL F L F S S A Y S R G V F R R D A H K S E V A H F f K D L G E ENFKjMVLIA F A Q Y L Q Q C P F 70 80 90 100 110 120 EDHVKLVNEV T E F A K T C V A D E S A E N C D K S L HTLFGDKLCT V A T L R E T Y G E M A D C C A K Q E P 130 2038Da 140 150 160 i097Da 170 180 E R N E C F L Q H K DDNPNLPR|LV R P E V D V M C T A FHDNEETFLK 210 KYLYEIARRH PYFYAPELLF 190 200 220 230 1061 Da 240 F A K R Y K A A F T E C C Q A A D K A A C L L P K L D E L R D E G K A S S A K Q R L K C A S L Q K F G E R A F K A W A V 250 1294 Da 260 i692 pa 270 280 290 300 ARLS SQRFPMA EFAEVSKLVT DLTKK/HTECC HGDLLECADD RADLAKYICE NQDSISSKLK 310 320 330 340 350 360 E C C E K P L L E K SHCIAEVEND EM PAD LPS LA A D F V E S K D V C K N Y A E A K D V F L G M F L Y E Y A R 370 1338 Da 380 390 400 410 420 R H P D Y S W L L LRJLAKTYETT L E K C C A A A D P H E C Y A K V F D E F K P L V E E P Q N LIKQNCELFE 430 440 2116 Da 450 460 470 480 Q L G E Y K F Q N A LLVFjYTKKVP QVSTPTLVEV SRfMLGKVGSK C C K H P E A K R M P C A E D Y L S W 490 500 510 520 530 540 LNQLCVLHEK f P V S D R V T K C C T E S L V N R R P C F S A L E V D E T Y V P K E F N A E T FTFHADICTL 550 1170 Da 560 570 580 590 600 SEKERQIWKQ T A L V E L V K H K P K A T K E Q L K A V M D D F A A F V E K C C K A D D K E T C F A E E G M K L V 1183 Da 609 A A S Q A A L G L 69 possible to distinguish regular, unaltered lysines from their modified counterparts. An example of this are the two overlapping peptides obtained from 247 - 264 in the HSA sequence. For the peptide at 1294Da, Lys249 is acetylated whereas Lys257 is not, making it a site for tryptic cleavage. The opposite scenario is true for the peptide at 1692Da where Lys249 serves as a cleavage site and Lys257 is acetylated. Such variable cleavage can be noticed throughout the sequence since most peptides are the result of unmodified lysines before their N-termini or at their C-terminal ends. In total, 10 lysine-acetylated peptides could be detected using the Q-Trap at various intensities. 4.2.2 Peptide Sequencing and the Q-Trap The Q-Trap is able to combine all the standard scan types of a typical triple quadrupole instrument with the sensitivity of an ion trap instrument. In addition, it is able to perform several MS scans sequentially from which peptides are selected for sequencing. Because we were interested in the analysis of post-translationally modified proteins, a survey scan dedicated to the detection of a particular modification was combined with a comprehensive MS scan which detected all peptides including the unmodified ones for protein identification purposes. The PREC scan is able to reliably detect stable post-translational modifications on peptides while the E M C scan is a sensitive MS scan capable of trapping and accumulating and scanning out peptides possessing more than one charge from Q3. Thus, these were selected as the two sequential survey scans from which peptides would be selected for MS/MS. 70 The selection process of peptides in two sequential survey scans is based on their intensities in either scan and is software-governed by Information Dependent Acquisition (IDA) as shown in Figure 4.2. The m/z values of peptides present in either of the MS scans along with their respective intensities are combined and added to a general list should they meet a minimum intensity threshold. Peptides are removed from this list by the software i f they specifically appear on a separate exclusion list which was defined before the analysis or i f they were added to the exclusion list temporarily after MS/MS to prevent repeated fragmentation. Redundant m/z values which are present in both scans are also removed and all signals are sorted based on their intensities. We set the software to select the top 1 - 3 most intense peptides for sequencing prior to the acquisition of another round of survey scans. After this selection process is complete, ions first undergo an "Enhanced Resolution" (ER) scan which singles out a peptide and traps it in Q3 for increased sensitivity. The ion is then scanned out at 250Th/s allowing the exact mass and charge state to be deduced. This information is then used for the final "Enhanced Product Ion" (EPI) scan where the precursor ion is subjected to collision induced dissociation (CID) and its fragments trapped and scanned out of Q3. The MS/MS spectrum of the fragments can then be submitted to a database search for protein identification. 4.2.3 Combined Survey Scans With the above system configuration, a digest mixture containing a 1:1 ratio of chemically acetylated HSA with the Mark 12 wide range molecular weight 71 Figure 4.2 Q-Trap Workflow. Schematic workflow for peptide sequencing using two survey scans in parallel. The Enhanced Multiply Charged (EMC) and the Precursor ion (PREC) scans were used as survey scans from which peptides would be selected by Information Dependent Acquisition (IDA). Peptides selected by IDA are then scanned again using an Enhanced Resolution (ER) scan followed by an Enhanced Product Ion (EPI) scan for sequence information. This whole process is completed prior to another round of survey scans. Survey 1 (eg. EMC) • Survey 2 [ (eg. PREC) Most intense peak(s) selected i Enhanced Resolution y J f MS/MS (EPI) Sequence information for most intense peaks only 72 standards was tested. The unmodified standards were used to simulate a biologically relevant situation, such as an immunoprecipitation where one would have a protein of interest (ie. acetylated HSA) along with background or interacting proteins which complicate the analysis. During the analysis, most of the peptides obtained were unmodified as expected with some modified peptides indicating that acetylated peptides were indeed being selected for MS/MS by IDA. Of the 10 modified peptides present in the HSA digest, only 7 were selected by IDA and were therefore sequenced using the combination of E M C and PREC as survey scans. Surprisingly and unfortunately, the 3 modified peptides missed by IDA eluted together (Fig. 4.3). Upon examination of the time of elution, the PREC scan (Fig. 4.3B) clearly showed 3 ions which produced a fragment ion of 126.1Th along with some minor noise near the baseline. These same peptides were also present in the E M C scan (Fig. 4.3A), however there were a number of other unmodified ions that do not appear in the PREC scan which also co-elute. Even though these ions are present in the PREC scan with intensities exceeding the required minimum threshold for IDA selection, none were submitted for MS/MS analysis. A closer look at some related factors was necessary including i f the PREC scan alone was indeed sensitive enough to ensure all ions could make it onto the IDA list in the first place. Furthermore, i f this scan proved to be sufficient in this regard, would other contributing factors such as the intensities of the individual survey scans be responsible for non-sequenced modified peptides? First, the PREC scan was tested to rule out a possible loss in sensitivity from the sequential survey scans. 73 Figure 4.3 Combining the survey scans: E M C + P R E C . E M C scan (A) and a PREC scan (B) at 39.0min in an experiment where both were used in parallel as survey scans. Co-eluting modified peptides at 549.8Th, 586.5Th and 635.5Th are shown in the PREC scan directly below their equivalents at 549.4Th, 585.8Th and 634.8Th in the E M C scan. 1582.4 B 687.2 I .i.u, L l i . M till, 790.7 iijll.niti 540 580 620 660 700 740 780 820 860 900 m/z 540 580 620 660 700 m/z 740 780 820 860 900 74 4.2.4 P R E C Sensitivity The sensitivity of this scanning mode was tested for its limit of detection using a dilution series of the peptide mixture analyzed in the previous combined survey scan experiment. The PREC scan alone was tested to see i f it alone could trigger IDA for all modified peptides, especially those which could not be sequenced using the combined scan mode. When the same amount of test digest (lOOOfmol) used in the combined scan mode was analyzed with the PREC scan by itself, all modified peptides were fragmented indicating that the intensities of these peptides were sufficient to trigger IDA. Even diluting the digest concentration to one half and loading 500fmol on column, all modified peptides were still able to trigger MS/MS scans. Only after a further dilution where only 333fmol was loaded were only 5 of the 10 modified peptides below the required threshold for IDA selection and were thus not selected. To further prove that this effect was dependent on the amount of digest used, the digest was diluted even further to where only 250fmol was loaded on column. In this experiment only 2 of the 10 acetylated peptides were accounted for on the IDA list. The intensities for extracted ion chromatograms (XICs) of three peptides at the various amounts of digest used for PREC scan only are shown in Figure 4.4. As shown, the intensities gradually decrease with a decrease in the amount of digest loaded on column proving that below a certain concentration, some modified peptides fell below the required threshold. Once this occurred, IDA no longer considered them even though they were clearly still detectable. Thus, they were not submitted for MS/MS and no sequence information was provided. This series of runs proved that for the modified peptides missed when combining E M C and PREC survey scans, the concentration needed for both detection and for 75 satisfying the threshold requirement was significantly lower than that which was used. However, even though the sensitivity of the PREC scan is sufficient to detect all modified peptides, optimizing the scan such that it is operating at its highest sensitivity possible is a worthwhile task to completely eliminate any possible PREC scan shortcomings. One possible way of gaining sensitivity in the PREC scan was to optimize a parameter known as the dwell time which dictates the amount of time spent on a particular m/z range. The dwell time is simply calculated by the total acquisition time for the scan divided by the number of steps or individual m/z ranges which are scanned. In theory, increasing the time the spectrometer spends or dwells at a particular mass range should increase sensitivity since more time is spent measuring a certain ion and thus more ions reach the detector. The aim of this optimization was to determine if, at a concentration where some modified peptides were no longer selected for IDA due to insufficient intensities, the number of these modified peptides could be increased with an increase in dwell time. Thus we loaded 333fmol of digest on-column since this was an amount where IDA only selected 5 of the 10 modified peptides and increased.the dwell time to determine i f an increase in the number of modified peptides selected by IDA would result. When the dwell time of 0.625ms was increased by 50% to 0.938ms, the average total number of modified peptides selected for fragmentation in 3 consecutive experiments increased only slightly (Fig 4.5). Further increases of dwell time to 1.25 and 2.50ms resulted in no 76 Figure 4.4 Modified Peptide Detection and Digest Load. Extracted ion chromatograms (XICs) of peptides detected at (A) 531Th, (B) 634Th and (C) 585Th with the amount of digest loaded on column ranging from lOOOfmol down to 250fmol. i i 6 x & 5 o ^ 4 in I 3 c ~ 2 1 0 B 3.0 0 2.4 + LU X ps) 1.8 _o .& •55 1.2 c a> c 0.6 0.0 1.5 2" x IT 8; 0.9 g0.6 0.3 0.0 1000fmol 500fmol iuukJikjJiK.//—iiitu-k-.rJ. JWHUM1I_IU ii-42 / / 42 / / 333fmol 1000fmol Time (min) 42 + 250fmol 500fmol 333fmol f 1 39 / / 39 Time (min) lOOOfmol 500fmol 333fmol 39 / / 39 // 39 Time (min) 42 250fmol HULL , 39 250fmol 39 77 additional peptides. In fact, the total number of peptides seemed to decrease as the dwell time increased past 0.938ms (Fig 4.5). Thus past a certain dwell time, increasing it is not an effective means of boosting PREC scan sensitivity. Examining the results in Figure 4.5 more closely shows that when taking all the overlapping error bars into consideration, no significant increase in sensitivity was observed at all. The failure of the increased dwell time to significantly increase the sensitivity for individual modified peptides can be rationalized by the manner in which ion intensities are registered on the QTrap. Because intensities are calculated as counts per second (cps), scanning for a longer period of time does not affect the ion flux or the rate at which ions arrive at the detector. Thus sensitivity is not increased i f more time is spent at a certain m/z, as in the case of increasing the dwell time in the PREC scan. Thus, the total intensity of an ion does increase and modified peptides which were not intense enough to make the IDA list before the increase in dwell time remain unchanged with respect to intensity. 4.2.5 E M C Detuning After proving that the PREC scan alone is able to easily detect and trigger IDA-dependent MS/MS for all modified peptides at lOOOfmol loaded on-column through sensitivity tests and dwell time optimization, another possible explanation for the missed modified peptides in the dual survey scans experiment can be that an IDA bias exists for one of the scan types, namely the E M C scan. Since this scan uses trapping in Q3 where ions can be stored for a predefined period of time for increased sensitivity and signal intensity, it could be much more intense that the PREC scan where trapping is not 78 Figure 4.5 Increased P R E C Dwell Time. Number of modified peptides detected with increased PREC dwell time and 333fmol of digest loaded on-column. The dwell time of 0.625ms was increased up to 2.5ms and showed no significant increase in MS/MS sequencing of modified peptides. 10-•D a> Q 6 in a> T3 £ 5 a> a. •a 4 0) 0.625 0.938 1.250 Dwell Time (ms) 2.500 79 available. To assess this, some representative peptides from both combined scans (Fig. 4.6) including two of the three missed modified peptides were examined more closely to see i f such an intensity difference was considerable. Extracted Ion Chromatograms (XICs) for unmodified peptides at 582 and 790Th and modified peptides at 549 and 634Th are shown in Figure 4.6. XICs show the range(s) within a run where individual peptides elute as well as their respective total intensities. For a certain peptide at a particular charge state, a single, sharply defined XIC peak should be present, however other ions may also possess the same m/z value making several peaks possible within the same XIC (Fig 4.6D). For the peptides selected for closer examination, the XICs confirm that all co-elute within approximately 30sec of each other. The modified peptides at 549 and 634Th both have XICs in E M C and PREC scans since they are detected by both survey scan types. The intensity of the peptide at 549Th in the E M C (Fig, 4.6C) is roughly 100-fold higher than its equivalent in the PREC (Fig. 4.6E) scan. The same holds true for the intensities in the E M C (Fig. 4.6D) and PREC (Fig. 4.6F) mode of the other missed modified peptide at 634Th. It is clear that since IDA selects ions based solely on their intensities above a certain threshold, ions appearing in the PREC scan are no competition for ions present with E M C mode due to much lower signal intensities in the PREC scan. This argument assumes that there are more than 3 signals present in the IDA list for MS/MS since, the top 3 m/z values are selected. If there are only 3, then all would be selected for MS/MS. Thus, it stands to reason that modified peptides will only be selected and sequenced should their intensities exceed any co-eluting unmodified peptides in the E M C scan only or i f their intensities are 80 Figure 4.6 Modified and Unmodified Peptides. Extracted ion chromatograms (XICs) are shown for unmodified peptides at (A) 582Th and (B) 790Th along with modified peptides at (C) 549Th and (D) 634Th in E M C mode. XICs of the modified peptides are also shown for 549Th and 634Th in PREC mode (E + F). Peptides with similar m/z values which co-elute with the 634Th peptide in the E M C scan (D) are shown in gray. A EMC 582 39.40 1.4e7 1.2e7 1.0e7 5 10 15 20 25 30 35 40 45 50 55 Time (min) 38.93 •3" 8.0e6 in 0 6.0e6 4.0o6 2.0e6 0 c EMC 549 E PREC 549 15 20 25 30 35 40 Time (min) 38.88 3.6e6 3.2e6 "3T 2.8e6 Q. 3. 2.4e6 j& 2.0e6 '55 C 1-6e6 a •£ 1.2e6 O.OeS 4.0e5 0.0 It) 2.0e4 a. o ~ 1.604 e <2 1.2e4 s C 8000 I il 'II II inn i i n 15 20 25 30 35 40 45 50 Time (min) B EMC 790 138.51 D EMC 634 10 15 20 25 30 35 40 45 50 55 Time (min) 38.92 34.091 40,75 I 54.42 15 20 25 30 35 Time (min) 40 45 50 55 F PREC 634 38.93 I i i m u m l l i l l l .111 . 15 20 25 30 35 40 45 50 Time (min) 81 highest in the PREC scan which is unlikely in the present situation. Unfortunately, in this case there were several co-eluting peptides during this run, some of which were of higher intensity than the mentioned modified peptides. The peptides at 582 and 790Th co-eluted with the modified ones and signals which were roughly 4-fold more intense (Fig. 4.6A + 4.6B) and so since their intensities were greater, IDA selected them for MS/MS and missed the weaker modified peptides. One way of addressing this issue would be to create a scenario where both survey scans have equal likelihood of triggering IDA-dependent MS/MS. Thus, the intensity of the E M C scan would need to be lowered significantly to roughly the same order of magnitude as the PREC scan. Because the PREC scan was optimized to be as sensitive as possible for modified peptide detection, the next step is to detune the E M C scan. A series of experiments were done using the E M C as the only survey scan to reduce its intensity to allow PREC triggering of IDA, i f possible. Because the E M C uses trapping in Q3, a simple way of decreasing its overall sensitivity and the signal intensity of its ions would be to decrease the time spent trapping and accumulating them. The typical trap fill time is 50ms, but it can be manually adjusted down to a minimum of 0.01ms. We first decreased the trap fill time from 50ms to 5ms and then down to the absolute minimum setting of 0.01ms. These results were then plotted using the intensities for 3 unmodified and 3 modified peptides (Fig. 4.7) to gauge the signal attenuation. When the trap fill time was reduced from 50 to 5ms, the intensities for the representative unmodified and modified peptides did not decrease significantly in most 82 cases. Thus, a decrease of 10-fold in fill time from 50 to 5ms did not translate into a similar decrease for individual ion intensities (Fig. 4.7). Further decreasing the trap fill time by 500-fold from 5 to 0.01ms did not decrease the overall ion intensities to the same extent. In fact, for most of the ions, the slight decrease was completely insignificant when taking error bars into account. Since the aim of decreasing the trap fill time was to reduce E M C signal intensities, we concluded that decreasing the trap fill time was not an effective means of balancing the overall intensities between the two survey scans and would not affect the IDA bias towards the E M C mode. Several explanations are possible for the insignificant decrease in ion strength with unusually low trap fill time: 1. The concentration of the sample was much too high and caused trap saturation no matter what fill time was used. With instruments which use ion trapping, a noticeable phenomenon known as "space-charge effects" occurs during ion trap saturation. This effect is the result of too many ions in close physical proximity which distorts their electric fields and impairs performance of the trap. A tell-tale sign of the space-charge effect is peak splitting of ions causing decreased resolution and mass accuracy. Neither of these effects were noticed in the spectra. Moreover, the counts registered in the spectra were not near the saturation of the detector and thus, trap saturation can be excluded as a possible explanation. 83 Figure 4.7 E M C Trap F i l l Time and Signal Intensity. XIC intensities for three ions at 531Th, 647Th, and 670Th in E M C mode at different trap fill times ranging from 50ms down to 0.01ms. The decrease in ion intensity is not proportional to the decrease in trap fill time and does fall to an intensity comparable to average PREC intensities for individual ions (~104). 8 50 5 0.01 50 5 0.01 50 5 0.01 Trap Fill Times (ms) 84 2. During the E M C , a quadrupole before QI known as QO is used for trapping while ions are being accumulated and scanned out of Q3. This adds more sensitivity and also increases the duty cycle while Q3 is closed during ion ejection. The trapping in QO can be toggled and, in this case, was turned off to negate all possible trapping elsewhere in the instrument. Therefore no other means of ion accumulation in the instrument could compensate the decrease in Q3 trapping time and could also be excluded as an explanation for the only slight decreases in intensities observed. Thus trap saturation and alternative trapping do not provide any insights into the lack of signal reduction when E M C trap fill times are decreased significantly. The manner by which the instrument interprets detected ions may explain the insignificant decrease of signal intensity with trap fill time. Since ions are detected as counts per second, the total scan time of 250ms for the E M C scan compensates for the reduction of trap fill times from 50 down to 0.01ms. Thus an E M C scan with 50ms trap fill time would last for 300ms whereas 0.01ms would last for 250.01ms. Since ions counts are normalized for every second, the total scan times for highest and lowest fill times are comparable and so the numbers of ions counted per second do not decrease significantly. 4.2.6 E M S and Dynamic F i l l Time The enhanced MS (EMS) scan is identical to the E M C scan, except that it does not eliminate singly-charged ions like the E M C but rather accepts all ions. The EMS scan has an added feature not available in the E M C scan known as "dynamic fill time" (DFT) which allows it to adjust Q3 trap fill time on-the-fly. By adjusting the trap fill 85 times dynamically, this ensures that data quality is optimized over a wide range of analyte concentrations. Ions which are very abundant will have a short fill time to reduce the possibility of space-charge effects by limiting their amounts. On the other hand, low abundant ions would be accumulated in the trap for longer to increase their intensities. Also, DFT is able to adjust the overall intensity of the EMS scan according to a "suggested" trap fill time entered by the user which proportionately increases or decreases counts depending on the time entered. Thus, DFT can serve as an attenuation parameter for ion signal intensity. To cause signal reduction, DFT is able to mathematically scale down the intensity proportionately with the suggested trap fill time entered by the user. Because this is simply a mathematical adjustment to the data, there are no adverse effects on ion detection and data quality such that signal-to-noise ratio kept constant while the intensity of a particular signal is reduced. Thus, we chose to investigate the applicability of the EMS survey scan using DFT and its effect on signal strength reduction. Trap fill times identical to the ones used for E M C scans were chosen for scans involving the EMS scan with DFT activated. The intensities for ions at 531Th, 647Th, and 670Th with 50, 5 and 0.01ms DFT fill times in EMS mode using DFT are shown in Table 2. Using DFT, the reduction in ion intensity is proportional with the drop in fill time. Thus, reducing the DFT fill time by one order of magnitude from 50 to 5ms reduces the signal of individual ions by 10-fold. Further decreasing the DFT fill time from 5ms to the minimum setting of 0.01ms causes a decrease of 500-fold. In fact, intensities for all ions at 0.01ms were in the hundreds compared to 105—106 with 50ms fill times. Thus the decrease of trap fill time is largely proportional to the decreases in signal strength. Also shown in Table 2 is the noise 86 Table 2 E M S DFT Trap Fill Times and Intensity. A comparison of individual ion intensities for three peptides at 531Th, 647Th and 670Th in EMS mode using three different trap fill times with dynamic fill time (DFT) activated. The intensities for the noise associated with each peptide as well as their signal to noise ratios are also shown. m/z Trap Fill Time (ms) Signal (cps) Noise (cps) Signal/Noise 531 . 50 1.9x106 ± 3.5x105 1.1x105 ± 4.3x104 17 ±7 5 2.2x105± 7.1x103 1.2x104±707 18 ±2 0.01 375 ±35 17.5 ±6.4 21 ±8 647 50 1.2x106 ± 4.9x105 1:3x10s ± 4.9x104 9 ±4 5 1.3x105± 7.1x103 9.9x103± 1.6x103 13 ±2 0.01 190 ±42 17.5 + 3.5 11 ±3 670 5 0 3.6x105 ± 6.4x104 2.4x104 ± 2.1x103 15 ±3 5 2.7x104 ± 6.4x103 2.3x10 3±70 12 ±3 0.01 80 ±7 4.3 ±0.4 19 ±3 87 associated with each of the signals. While the noise was typically 1-2 orders of magnitude lower than the signals, it also displayed the similar decrease in signal intensity as the DFT trap fill time is lowered. As the trap fill time was lowered by an order of magnitude, so was the noise intensity. Equipped with both the signal and noise intensities, the signal/noise (S/N) ratio was calculated and, as shown in Table 2, remains relatively constant within errors. The average S/N values for individual peptides were 18 for 531Th, 11 for 647Th, and 15 for 670Th and varied slightly with fill time. The lowest setting of 0.01ms is actually much too low causing most ions to never be selected by IDA since they now fail to meet the previously set threshold of 5000 counts. This threshold was initially set to account for the noise in the PREC scan and remained constant throughout all experiments for consistency. Thus, the trap fill time needed to be optimized so that the signal was comparable to the average PREC peptide intensity (~104) in the EMS scan. After some experimentation, a dynamic fill time of 1.5ms using DFT in EMS mode created individual ion intensities on the order of those expected in the PREC scan. Since the intensities were equalized for both survey scans, the original PREC and EMS with DFT activated scans were combined in order to assess i f more modified peptides could be selected by IDA and sequenced. This would be compared to the first experiment where the PREC scan was combined with the E M C scan and caused IDA to miss 3 of the 10 modified peptides. 88 4.2.7 Combining Optimized E M S with P R E C When combining the original PREC scan with the optimized EMS scan with active DFT where both scans contain ions of nearly equal average signal intensities, the number of modified peptides submitted for MS/MS does not increase. In fact, the number of sequenced modified peptides decreased from 7 of 10 in the original EMC/PREC experiment (Fig 4.3) to 6 when using the original PREC scan with the intensity-adjusted EMS scan. A n example of where peptides are lost is shown in Figure 4.8A of the EMS scan at 42min. Three modified peptides are present in this scan as confirmed by the PREC scan (Fig 4.8B). The intensities of all three modified peptides at 549Th, 586Th, and 635Th are almost exactly equal in both the EMS (Fig. 4.8A) and the PREC (Fig. 4.8B) scans. The acetylated peptide at 549Th in this set is submitted for IDA analysis, however the other two aren't. The weak 586Th peptide fails to meet the required 5000 count IDA threshold in both cases and stands no chance of MS/MS admission. The 634Th ion, however, is clearly above this threshold in both cases, but is not sequenced. Curiously, its intensity is higher than the 549Th ion which falls below the threshold in the PREC scan. Upon close examination, however, the intensity of the 549Th ion is slightly higher than the 634Th ion in the EMS scan. Thus, the three ions which were submitted from both sets of peptides from both survey scans were in the order of the strongest detected ions: 523.4Th, 655.4Th, and 549.5Th. Thus, even though intensities of the individual ions were matched in both survey scans, information was still lost due to chromatographic separation. The fact that one less peptide was sequenced with the optimized settings was due to chromatographic variation between runs. 89 Figure 4.8 Combining the survey scans: P R E C + Optimized E M S with DFT. Combined EMS with DFT (A) and PREC (B) scans at 42min within the same run. Acetylated peptides are at 549Th, 586Th, and 635Th. 10.01 ? 7.5] L U &5.0I in | 2.51 0.0 uJjbJl 523.4 jkii 549.5 635.0 .586.1 655.4 M L l l M II I i,ill m 540 580 620 660 700 m/z 740 780 820 860 900 B 5 6 5.0 CO o + LU x 4.0 uT a o ^ 3.0 'Si c a> c 2.0 1.0 550.5 635.5 1587.3 540 580 620 660 700 740 780 820 860 900 m/z 90 4.2.8 Concluding Remarks Finding a way to decrease the overall intensity of the EMS scan to a level of the PPvEC scan still failed to sequence all 10 of the modified peptides in the complex mixture however this was possible when using the PREC scan alone. In this case, many peptides were found to co-elute causing only the most intense signals to be selected for sequencing and so i f modified peptides happened to be the most intense signals in the EMC/EMS scans, only then were they candidates for IDA selection. Otherwise, they were dominated by more intense co-eluting unmodified peptides. Thus, these scans should ideally be done separately in order to ensure that all modified peptides are selected by IDA in the PREC scan and that as many unmodified peptides as possible are sequenced in the EMC/EMS scan to provide confident protein identification. Since the scans would now be done separately, the analysis time and the amount of analyte needed to carry out both experiments would be doubled. Although not always possible, decreasing sample complexity could potentially bypass the difficulties encountered here and allow the user to obtain maximum unmodified peptide sequencing as well as cover all present post-translationally modified peptides within a single L C run. Alternatively, i f complexity cannot be decreased without the loss of valuable sample, stretching the L C gradient time would be another option. This would allow for better resolution and fewer co-eluting peptides in each scan and could minimize information loss due to sample complexity. The sample complexity, although known in this study, is normally unknown with real biological samples. Thus, the extent to which the L C gradient times should be extended could not be estimated 91 accurately. Moreover, L C gradients could be extended for very long times which may lead to impractically long analysis times. The real issue, however, is the "non-biased" use of peptide intensities which causes the problem of modified peptides missed by IDA. If one survey scan, such as the PREC scan in this case, had priority over the other survey scan, then as soon as signals over the threshold were detected in the PREC scan they would be selected and submitted for MS/MS. Thus even i f there are 3 co-eluting modified peptides, as seen earlier, then they would all be sequenced successfully using this type of setup. If no signals are present in the PREC scan, then the software would automatically select the 3 most intense signals from the E M C scan for sequencing. Using this approach, all modified peptides would be sequenced along with many, unmodified peptides for protein identification. Unfortunately, this type of setup is currently unavailable with QTrap software. 92 C H A P T E R 5: G E N E R A L C O N C L U S I O N S 5.1 Summary Current methods for intact protein isolation from SDS-PAGE gels include electro-elution and passive diffusion. An alternate approach was described in Chapter 3, where a dissolvable SDS-PAGE gel system was created and successfully applied for the isolation of several proteins ranging from 17 - 47kDa. In addition to not only being able to isolate proteins from these gels, other analytical techniques were also possible once the entire protein was purified. Mass spectrometry was used to obtain the exact molecular weight of one of the isolated proteins, a-casein, to convincingly show that 8 sites of phosphorylation were occupied by measuring the intact masses before and after treatment with a phosphatase. Another advantage of having intact protein is that the digestion efficiency and/or peptide recovery were increased compared to the traditional in-gel digestion procedure. In the case of chymotrypsin, the numbers of peptides recovered between the in-solution and the in-gel digestions were dramatic. Since post-translational modifications of proteins are an essential aspect of protein characterization, the only way of guaranteeing that modification-containing peptides are obtained from proteins found within SDS-PAGE gels is to isolate the intact protein. Even when ensuring that modification-containing peptides are obtained by isolation of the intact proteins, detection and sequencing of these important peptides is yet another hurdle for full protein characterization. The recently introduced hybrid triple quadrupole/linear ion trap (QTrap, MDS Sciex/Applied Biosystems) instrument has the 93 capability of detecting and obtaining sequencing information for both modified and unmodified peptides by combining two survey scans in parallel within the same run. Unfortunately, our preliminary data using this type of setup showed that some modified peptides were not sequenced due to a sequencing bias towards the general survey scan which detected both modified and unmodified peptides. Realizing that the cause of the bias was intensity-based, this lead us through a series of experiments aimed to ease this bias and to allow for the modification-specific survey scan to trigger sequencing of its peptides. Even after optimizing both survey scans to alleviate this problem, modified peptides were still missed when using both scan types within the same L C run. Thus, due to sample complexity issues, we concluded that making use of parallel scans on the QTrap should only be used with care as valuable information may be lost otherwise. 5.2 Future Directions Along with all the advantages of isolating intact proteins from dissolvable gels, there are also some drawbacks or limitations, as with any method in development. The first major drawback is that reducing SDS-PAGE cannot be performed due to the sensitivity of the B A C crosslinker to dithiothreitol and P-mercaptoethanol. Thus i f sample reduction is desired for future steps such as digestion after electrophoresis, reduction and alkylation would need to be done prior to loading. This step would be necessary for proteins possessing disulfide linkages which would include many proteins. Another technical consideration would be the solubility of proteins during the polyacrylamide precipitation step using isopropanol. Proteins tend to precipitate at 94 higher concentrations of organic solvent and may experience different degrees of solubility during this step even though the minimal amount of isopropanol is used in addition to HFIP. Removing dissolved polyacrylamide matrix is an essential step prior to protein purification by precipitation with acetone. If the polyacrylamide is not removed during this step, it will precipitate during the acetone precipitation along with the protein and cause re-solubilization difficulties. Future work investigating gentler approaches for dissolved polyacrylamide removal would be well worthwhile as this would minimize sample loss and provide more consistent recoveries of a variety of proteins. To obtain an accurate intact mass measurement on a particular protein, it must ideally be the only protein being analyzed. If there are several proteins or even many isoforms of a single protein within an excised SDS-PAGE band, an accurate mass measurement will not be possible due to signal interference from multiply-charged protein signals. If a tissue extract was run on a 1-D BAC-crosslinked gel, several proteins of similar molecular weight would likely co-migrate as a single band and result in a mixture of proteins after purification from the excised band. Thus using B A C -crosslinked gel as the second dimension in a 2-D gel would be the next logical step to increase resolution of protein isoforms and allow whole cell lysate or at least cell fractions to be run without concerns for protein co-purification. A 2-D gel would allow for the exploration of the full potential of BAC-crosslinked gel since individual protein isoforms could be analyzed by mass spectrometry to determine their modification status or for deducing splice variants. 95 R E F E R E N C E S Allfrey, V . G . ; Faulkner, R.; Mirsky, A .E . Acetylation and methylation of histones and their possible role in the regulation of R N A synthesis. Proc Natl Acad Sci U.S.A. 1964. 51; 786-94. , Anker, H.S. A solubilizable acrylamide gel for electrophoresis. Febs Lett. 1970. 7; 293. Apweiler, R.; Hermjakob, H. ; Sharon, N . On the frequency of protein glycosylation, as deduced from analysis of the SWISS-PROT database. Biochim Biophys Acta. 1999. 1473; 4-8. Babu, C.V.; Lho, D.S.; Yoo, Y.S. Monitoring the neuropeptide metabolites by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. J Pharm Biomed Anal. 2006. 40; 136-41. Belew, M . ; Yip, T.T.; Andersson, L. ; Ehrnstrom, R. High-performance analytical applications of immobilized metal ion affinity chromatography. Anal Biochem. 1987. 164; 457-65. . Bemdt, P.; Hobohm, U . ; Langen, H. Reliable automatic protein identification from matrix-assisted laser desorption/ionization mass spectrometric peptide fingerprints. Electrophoresis. 1999. 20; 3521-3526. Bhaskar, S.; Dutt, S.; Mukherjee, R. A simple method of electroelution of individual protein bands from SDS polyacrylamide gels for direct study in cellular assays. J Immunoassay. 2000.21; 355-375. Boja, E.S.; Fales, H . M . Overalkylation of a protein digest with iodoacetamide. Anal Chem. 2001. 73; 3576-82. Bolbach, G. Matrix-assisted laser desorption/ionization analysis of non-covalent complexes: fundamentals and applications. Curr Pharm Des. 2005. 11; 2535-57. Bongertz, V . Elution of proteins from polyacrylamide gel: description of a simple and economic method. Rev Inst Med Trop. 1989. 31; 44-47. Boschero, A .C . ; Negreiros de Paiva, C E . Transplantation of islets of Langerhans in diabetic rats. Acta Physiol Lat Am. 1977. 27; 1-6. Bunai, K.; Yamane, K . Effectiveness and limitation of two-dimensional gel electrophoresis in bacterial membrane protein proteomics and perspectives. J Chromatogr B Analyt Technol Biomed Life Sci. 2005. 815; 227-36. Calatayud, C.A.; Pasquini, L .A . ; Pasquini, J .M.; Soto, E.F. Involvement of the ubiquitin-mediated proteolytic system in the signaling pathway induced by ceramide in primary astrocyte cultures. Dev Neurosci. 2005. 27; 397-407. 96 Calissano, M . ; Faulkes, D.; Latchman, D.S. Phosphorylation of the Brn-3a transcription factor is modulated during differentiation and regulates its functional activity. Brain Res Mol Brain Res. 2005.141; 10-8. Celis, J.E.; Gromov, P.; Ostergaard, M . ; Madsen, P.; Honore, B.; Dejgaard, K. ; Olsen, E.; Vorum, H. ; Kristensen, D.B.; Gromova, I.; Haunso, A . ; Van Damme, J.; Puype, M . ; Vandekerckhove, J.; Rasmussen, H.H. Human 2-D P A G E databases for proteome analysis in health and disease: FEBS Lett. 1996. 398; 129-134. Chang, H.T.; Yergey, A . L . ; Chrambach, A . Electroelution of proteins from bands in gel electrophoresis without gel sectioning for the purpose of protein transfer into mass spectrometry: elements of a new procedure. Electrophoresis. 2001. 22; 394-398. Cheng, X . ; Cole, R.N. ; Zaia, J.; Hart, G.W. Alternative O-glycosylation/O-phosphorylation of the murine estrogen receptor beta. Biochemistry. 2000. 39; 11609-11620. Chiari, M . ; Righetti, P.G.; Negri, A . ; Ceciliani, F.; Ronchi, S. Preincubation with cysteine prevents modification of sulfhydryl groups in proteins by unreacted acrylamide in a gel. Electrophoresis. 1992. 13; 882-884. Choules, G.L.; Zimm, B .H. An acrylamide gel soluble in scintillation fluids: its application to electrophoresis at neutral and low pH. Anal. Biochem. 1965. 13; 336-344. Claverol, S.; Burlet-Schiltz, O.; Gairin, J.E.; Monsarrat, B. Characterization of protein variants and post-translational modifications: ESI-MSn analyses of intact proteins eluted from polyacrylamide gels. Mol Cell Proteomics. 2003. 2; 483-493. Cohen, P. The regulation of protein function by multisite phosphorylation~a 25 year update. Trends Biochem Sci. 2000. 25; 596-601. Cohen, S.L.; Chait, B.T. Mass spectrometry of whole proteins eluted from sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels. Anal Biochem. 1997. 247; 257-267. Corthals, G.L.; Gygi, S.P.; Aebersold, R.; Patterson, S.D. Identification of proteins by mass spectrometry. In: Rabilloud (Ed.). Proteome Research: Two-Dimensional Gel Electrophoresis and Identification Methods, Springer-Verlag, Berlin, Germany. 2000. pp. 197-231. Dauphinee, S.M.; Ma, M . ; Too, C.K. Role of O-linked beta-N-acetylglucosamine modification in the subcellular distribution of alpha4 phosphoprotein and Spl in rat lymphoma cells. J Cell Biochem. 2005. 96; 579-588. Davies, A.R.; Belsey, M.J. ; Kozlowski, R.Z. Volume-sensitive organic osmolyte/anion channels in cancer: novel approaches to studying channel modulation employing proteomics technologies. Ann N YAcad Sci. 2004. 1028; 38-55. 97 Davis, M.T.; Beierle, J.; Bures, E.T.; McGinley, M.D. ; Mort, J.; Robinson, J.H.; Spahr, C.S.; Yu, W.; Luethy, R.; Patterson, S.D. Automated L C - L C - M S - M S platform using binary ion-exchange and gradient reversed-phase chromatography for improved proteomic analyses. J Chromatogr B Biomed Sci Appl. 2001. 752; 281-91. DeKeyser, S.S.; L i , L. Matrix-assisted laser desorption/ionization Fourier transform mass spectrometry quantitation via in cell combination. Analyst. 2006. 131; 281 -90. DeLange, R.J.; Smith, E.L.; Bonner, J. Calf thymus histone 3: sequences of the amino-and carboxyl-terminal regions and of the regions containing lysyl residues modified by acetylation and methylation. Biochem Biophys Res Commun. 1970. 40; 989-93. Deschamps, S.; Viel , A . ; Garrigos, M . ; Denis, H. ; le Maire M . mRNP4, a major mRNA-binding protein from Xenopus oocytes is identical to transcription factor F R G Y2. J Biol Chem. 1992. 267; 13799-802. Dunn, M J . Electroelution of proteins from polyacrylamide gels. Methods Mol Biol. 2004. 244; 339-343. Ehring, H. ; Stromberg, S.; Tjernberg, A . ; Noren, B. Matrix-assisted laser desorption/ionization mass spectrometry of proteins extracted directly from sodium dodecyl sulphate-polyacrylamide gels. Rapid Commun Mass Spectrom. 1997. 11; 1867-1873. Faulkner, R.D.; Carraway, R.; Bhatnagar, Y . M . Preparative electrophoresis of histones in solubilizable polyacrylamide gels. Biochem. Biophys. Acta 1982. 708; 245-252. Fazili, K . M . ; Mir, M . M . ; Qasim, M . A . Changes in protein stability upon chemical modification of lysine residues of bovine serum albumin by different reagents. Biochem Mol Biol Int. 1993. 31; 807-816. Feick, R.G.; Shiozawa, J.A. A high-yield method for the isolation of hydrophobic proteins and peptides from polyacrylamide gels for protein sequencing. Anal Biochem. 1990. 187; 205-211. Fenselau, C. M A L D I MS and strategies for protein analysis. Anal Chem. 1997. 69; 661A-665A. Fernley, H.N.; Bisaz, S. Studies on alkaline phosphatase. Phosphorylation of calf-intestinal alkaline phosphatase by 32P-labelled pyrophosphate. Biochem J. 1968. 107; 279-83. Fichmann, J.; Westermeier, R. 2-D protein gel electrophoresis. A n overview, p. 1-7. In A.J . Link (Ed.). 2-D Proteome Analysis Protocol. 1999. Humana Press, Totowa, NJ. Figeys, D.; McBroom, L.D. ; Moran, M.F. Mass spectrometry for the study of protein-protein interactions. Methods. 2001. 24; 230-239. Flory, P.J. Principles of Polymer Chemistry; Cornell University Press: Ithaca, N Y , 1953. 98 Flyer, D . C ; Tevethia, S.S. Biology of Simian Virus 40 (SV40) transplantation antigen (TrAg). VIII. Retention of SV40 TrAg sites on purified SV40 large T antigen following denaturation with sodium dodecyl sulfate. Virology 1982. 117; 267-270. Garrigos, M . ; Deschamps, S.; Viel , A. ; Lund, S.; Champeil, P.; Moller, J.V.; le Maire, M . Detection of Ca(2+)-binding proteins by electrophoretic migration in the presence of Ca2+ combined with 45Ca2+ overlay of protein blots. Anal Biochem. 1991. 194; 82-8. Garzotti, M . ; Rovatti, L. ; Hamdan, M . On-line liquid chromatography/electrospray tandem mass spectrometry to investigate acrylamide adducts with cysteine residues: implications for polyacrylamide gel electrophoresis separations of proteins. Rapid Commun Mass Spectrom. 1998. 12; 484-488. Gasthuys, M . ; Alves, S.; Fenaille, F.; Tabet, J.C. Simple identification of a cross-linked hemoglobin by tandem mass spectrometry in human serum. Anal Chem. 2004. 76; 6628-34. Gelfi, C ; Alloni, A . ; De Besi, P.; Righetti, P.G. investigating the reaction of the properties of acrylamide bifunctional cross-linkers with P-lactoglobulin by matrix-assisted laser desorption/ionization-mass spectrometry. J. Chromat. 1992. 608; 343-348. Gevaert, K.; Demol, FL; Martens, L. ; Hoorelbeke, B.; Puype, M . ; Goethals, M . ; Van Damme, J.; De Boeck, S.; Vandekerckhove, J. Protein identification based on matrix assisted laser desorption/ionization-post source decay-mass spectrometry. Electrophoresis. 2001. 22; 1645-51. Gevaert, K. ; Vandekerckhove, J. Protein identification methods in proteomics. Electrophoresis 2000. 21; 1145-1154. Gomez, S.M.; Nishio, J.N.; Faull, K.F.; Whitelegge, J.P. The chloroplast grana proteome defined by intact mass measurements from liquid chromatography mass spectrometry. Mol Cell Proteomics. 2002. 1; 46-59. Goodlett, D.R.; Armstrong, F.B.; Creech, R.J.; van Breemen, R.B. Formylated peptides from cyanogen bromide digests identified by fast atom bombardment mass spectrometry. Anal Biochem. 1990. 186; 116-120. Goodman, D.; Matzura, H. An improved method of counting radioactive acrylamide gels. Anal. Biochem. 1981. 42; 481-486. Grazu, V. ; Ovsejevi, K. ; Cuadra, K. ; Betancor, L. ; Manta, C ; Batista-Viera, F. Solid-phase reducing agents as alternative for reducing disulfide bonds in proteins. Appl Biochem Biotechnol. 2003. 110; 23-32. Hager, J. A new linear ion trap mass spectrometer. Rapid Comm Mass Spectrom. 2002. 16; 512-26. Hancock, W.S.; Wu, S.L.; Shieh, P. The challenges of developing a sound proteomics strategy. Proteomics 2002. 2; 352-359. 99 Hansen, J.N. Chemical and electrophoretic properties of solubilizable disulfide gels. Anal. Biochem. 1980. 105; 192-201. Hansen, J.N. Electrophoresis of ribonucleic acid on a polyacrylamide gel which contains disulfide cross-linkages. Anal. Biochem. 1976. 76; 37-44. Hansen, J.N. Use of solubilizable acrylamide disulfide gels for isolation of D N A fragments suitable for sequence analysis. Anal. Biochem. 1981. 116; 146-151. Harris, S.L.; Levine, A . J . The p53 pathway: positive and negative feedback loops. Oncogene. 2005. 24; 2899-2908. Henzel, W.J.; Bi l led, T .M. ; Stults, J.T.; Wong, S .C; Grimley, C ; Watanabe, C. Identifying proteins from two-dimensional gels by molecular mass searching of peptide fragments in protein sequence databases. Proc Natl Acad Sci U S A. 1993. 90; 5011-5015. Herbert, B. Advances in protein solubilization for two-dimensional electrophoresis. Electrophoresis 1999. 20; 660-663. Higham, C.E.; Jaikaran, E.T.; Fraser, P.E.; Gross, M . ; Clark, A . Preparation of synthetic human islet amyloid polypeptide (IAPP) in a stable conformation to enable study of conversion to amyloid-like fibrils. FEBS Lett. 2000. 470; 55-60. Hillenkamp, F.; Karas, M . ; Beavis, R . C ; Chait, B.T. Matrix-assisted laser desorption/ionization mass spectrometry of biopolymers. Anal Chem. 1991. 63; 1193A-1203A. Hisano, N . ; Morikawa, N . ; Iwata, H. ; Ikada, Y . ; Entrapment of islets into reversible disulfide hydrogels. J. Biomed. Mater. Res. 1998. 40; 115-123. Hogan, J.M.; Pitteri, S.J.; McLuckey, S.A. Phosphorylation site identification via ion trap tandem mass spectrometry of whole protein and peptide ions: bovine alpha-crystallin A chain. Anal Chem. 2003. 75; 6509-16. Hopfgartner, G.; Varesio, E.; Tschappat, V . ; Grivet, C ; Bourgogne, E.; Leuthold, L .A . Triple quadrupole linear ion trap mass spectrometer for the analysis of small molecules and macromolecules. J Mass Spectrom. 2004. 39; 845-55. Hunter, T. The Croonian Lecture 1997. The phosphorylation of proteins on tyrosine: its role in cell growth and disease. Philos Trans R Soc Lond B Biol Sci. 1998. 353; 583-605. Huq, M.D.; Khan, S.A.; Park, S.W.; Wei, L .N . Mapping of phosphorylation sites of nuclear corepressor receptor interacting protein 140 by liquid chromatography-tandem mass spectroscopy. Proteomics. 2005. 5; 2157-66. Jagannadham, M . V . ; Nagaraj, R. Detection of peptides covalently modified with multiple fatty acids by MALDI-TOF mass spectrometry. JPept Res. 2005. 66; 94-100. Janusz, M . ; Niezgodka, M . ; Wieczorek, Z.; Lisowski, J. A mild method for the purification of guinea pig peritoneal macrophage Fc gamma receptors. Affinity 100 chromatography and elution of the receptor with reducing agents. J Immunol Methods. 1986. 86; 19-24. Jensen, O.N.; Mortensen, P.; Vorm, O.; Mann, M . Automation of matrix-assisted laser desorption/ionization mass spectrometry using fuzzy logic feedback control. Anal Chem. 1997. 69; 1706-1714. Jorgensen, C.S.; Jagd, M . ; Sorensen, B.K.; McGuire, J.; Barkholt, V. ; Hojrup, P.; Houen, G. Efficacy and compatibility with mass spectrometry of methods for elution of proteins from sodium dodecyl sulfate-polyacrylamide gels and polyvinyldifluoride membranes. Anal Biochem. 2004. 330; 87-97. Kange, R.; Selditz, U . ; Granberg, M . ; Lindberg, U . ; Ekstrand, G.; Ek, B.; Gustafsson, M . Comparison of different I M A C techniques used for enrichment of phosphorylated peptides. JBiomol Tech. 2005. 16; 91-103. Karas, M . ; Hillenkamp, F. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal Chem. 1988. 60; 2299-301. Keller, P.J.; Fiordalisi, J.J.; Berzat, A .C. ; Cox, A.D. Visual monitoring of post-translational lipid modifications using EGFP-GTPase probes in live cells. Methods. 2005. 371; 131-37 Kim, J.Y.; Kim, K.W.; Kwon, H.J.; Lee, D.W.; Yoo, J.S. Probing lysine acetylation with a modification-specific marker ion using high-performance liquid chromatography/electrospray-mass spectrometry with collision-induced dissociation. Anal Chem. 2002. 74; 5443-9. Klose, J.; Kobalz, U . Two-dimensional electrophoresis of proteins: an updated protocol and implications for a functional analysis of the genome. Electrophoresis. 1995. 16; 1034-59. Koll i , V.S.; Orlando, R. A new strategy for M A L D I on magnetic sector mass spectrometers with point detectors. Anal Chem. 1997. 69; 327-32. Komatsu, Y . ; Yukutake, Y . ; Yoshida, M . Four different clones of mouse anti-acetyllysine monoclonal antibodies having different recognition properties share a common immunoglobulin framework structure. J Immunol Methods. 2003. 272; 161-75. Kunz, S.; Rojek, J .M.; Kanagawa, M . ; Spiropoulou, C.F.; Barresi, R.; Campbell, K.P.; Oldstone, M.B. Posttranslational modification of alpha-dystroglycan, the cellular receptor for arenaviruses, by the glycosyltransferase L A R G E is critical for virus binding. J Virol. 2005. 79; 14282-96. Le Maire, M . ; Deschamps, S.; Moller, J.V.; Le Caer, J.P.; Rossier, J. Electrospray ionization mass spectrometry on hydrophobic peptides electroeluted from sodium dodecyl sulfate-polyacrylamide gel electrophoresis application to the topology of the sarcoplasmic reticulum Ca2+ ATPase. Anal. Biochem. 1993. 214; 50-54. 101 le Maire, M . ; Deschamps, S.; Moller, J.V.; Le Caer, J.P.; Rossier, J. Electrospray ionization mass spectrometry on hydrophobic peptides electroeluted from sodium dodecyl sulfate-polyacrylamide gel electrophoresis application to the topology of the sarcoplasmic reticulum Ca2+ ATPase. Anal. Biochem. 1993. 214; 50-57. Lichty, J.J.; Malecki, J.L.; Agnew, H.D. Michelson-Horowitz DJ, Tan S. Comparison of affinity tags for protein purification. Protein Expr Purif. 2005. 41; 98-105. Liebler, D.C. Proteomic approaches to characterize protein modifications: new tools to study the effects of environmental exposures. Environ Health Perspect. 2002. 110; 3-9. Lin, Z.; Crockett, D.K.; Lim, M.S.; Elenitoba-Johnson, K.S. High-throughput analysis of protein/peptide complexes by immunoprecipitation and automated L C - M S / M S . J Biomol Tech. 2003. 14; 149-55. 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; 676-82. Liu, H. ; Lin, D.; Yates, J.R. 3rd. Multidimensional separations for protein/peptide analysis in the post-genomic era. Biotechniques. 2002. 32; 898-911. Loughnan, M . ; Bond, T.; Atkins, A. ; Cuevas, J.; Adams, D.J.; Broxton, N . M . ; Livett, B.G.; Down, J.G.; Jones, A. ; Alewood, P.F.; Lewis, R.J. Alpha-conotoxin EpI, a novel sulfated peptide from Conus episcopatus that selectively targets neuronal nicotinic acetylcholine receptors. J Biol Chem. 1998. 273; 15667-74. Mann, M . ; Hendrickson, R.C.; Pandey, A . Analysis of proteins and proteomes by mass spectrometry. Annu Rev Biochem. 2001. 70; 437-73. Mauger, F.; Jaunay, O.; Chamblain, V . ; Reichert, F.; Bauer, K. ; Gut, I.G.; Gelfand, D.H. SNP genotyping using alkali cleavage of R N A / D N A chimeras and M A L D I time-of-flight mass spectrometry. Nucleic Acids Res. 2006. 34; el 8. O'Connell, P.B.H.; Brady, C S . Polyacrylamide gels with modified cross-linkages. Anal. Biochem. 1976. 76; 63-73. Oberst, A. ; Rossi, M . ; Salomoni, P.; Pandolfi, P.P.; Oren, M . ; Melino, G.; Bernassola, F. Regulation of the p73 protein stability and degradation. Biochem Biophys Res Commun. 2005. 331; 707-12. O'Farrell, P.H.; High resolution two-dimensional electrophoresis of proteins. J Biol Chem. 1975. 250; 4007-4021 Ogorzalek, R.R.; Mitchell, C ; Stevenson, T.I.; Martin, S.A.; Hines, W . M . ; Juhasz, P.; Patterson, D.H.; Peltier, J .M.; Loo, J.A.; Andrews, P.C. Sensitivity and mass accuracy for proteins analyzed directly from polyacrylamide gels: implications for proteome mapping. Electrophoresis. 1997. 18; 382-390. 102 Opiteck, G.J.; Lewis, K .C . ; Jorgenson, J.W.; Anderegg, R.J. Comprehensive on-line LC/LC/MS of proteins. Anal Chem. 1997. 69; 1518-24. Ostergaard, M . ; Wolf, H . ; Orntoft, T.F.; Celis, J.E. Psoriasin (S100A7): a putative urinary marker for the follow-up of patients with bladder squamous cell carcinomas. Electrophoresis. 1999. 20; 349-354. Page, M.J.; Amess, B.; Townsend, R.R.; Parekh, R.; Herath, A . ; Brusten, L. ; Zvelebil, M.J.; Stein, R.C.; Waterfield, M.D. ; Davies, S .C; O'Hare, M.J . Proteomic definition of normal human luminal and myoepithelial breast cells purified from reduction mammoplasties. Proc Natl Acad Sci USA. 1999. 96; 12589-12594. Pandey, A. ; Fernandez, M . M . ; Steen, H. ; Blagoev, B.; Nielsen, M . M . ; Roche, S.; Mann, M . ; Lodish, H.F. Identification of a novel immunoreceptor tyrosine-based activation motif-containing molecule, STAM2, by mass spectrometry and its involvement in growth factor and cytokine receptor signaling pathways. J Biol Chem. 2000. 275; 38633-9. Pandey, A. ; Mann, M . Proteomics to study genes and genomes. Nature. 2000. 405; 837-846. Pandey, A. ; Podtelejnikov, A . V . ; Blagoev, B.; Bustelo, X.R. ; Mann, M . ; Lodish, H.F. Analysis of receptor signaling pathways by mass spectrometry: identification of vav-2 as a substrate of the epidermal and platelet-derived growth factor receptors. Proc Natl Acad Sci USA. 2000. 97; 179-84. Pawson, T.; Scott, J.D.; Signaling through scaffold, anchoring, and adaptor proteins. Science. 1997. 278; 2075-80. Puig, O.; Caspary, F.; Rigaut, G.; Rutz, B.; Bouveret, E.; Bragado-Nilsson, E.; Wilm, M . ; Seraphin, B. The tandem affinity purification (TAP) method: a general procedure of protein complex purification. Methods. 2001. 24; 218-29. Radko, S.P.; Chang, H.T.; Zakharov, S.F.; Bezrukov, L . ; Yergey, A . L . ; Vieira, N.E.; Chrambach, A . Electroelution without gel sectioning of proteins from sodium dodecyl sulfate-polyacrylamide gel electrophoresis: fluorescent detection, recovery, isoelectric focusing and matrix assisted laser desorption/ionization-time of flight of the electroeluate. Electrophoresis. 2002. 23; 985-992. Rappsilber, J.; Siniossoglou, S.; Hurt, E . C ; Mann, M . A generic strategy to analyze the spatial organization of multi-protein complexes by cross-linking and mass spectrometry. Anal Chem. 2000. 72; 267-275. Reckard, C.R.; Ziegler, M . M . ; Naji, A. ; Galbut, D.; Barker, C.F. Physiologic and immunologic status of long-functioning transplanted pancreatic islets in rats. Surg. Forum. 1974. 25; 374-6. Riebeseel, K. ; Biedermann, E.; Loser, R.; Breiter, N . ; Hanselmann, R.; Mulhaupt, R.; Unger, C ; Kratz, F. Polyethylene glycol conjugates of methotrexate varying in their 103 molecular weight from M W 750 to M W 40000: synthesis, characterization, and structure-activity relationships in vitro and in vivo. Bioconjug Chem. 2002. 13; 773-85. Rigaut, G.; Shevchenko, A . ; Rutz, B.; Wilm, M . ; Mann, M . ; Seraphin, B. A generic protein purification method for protein complex characterization and proteome exploration. Nat Biotechnol. 1999. 17; 1030-2. Salek, M . ; Alonso, A . ; Pipkorn, R.; Lehmann, W.D. Analysis of protein tyrosine phosphorylation by nanoelectro spray ionization high-resolution tandem mass spectrometry and tyrosine-targeted product ion scanning. Anal Chem. 2003. 75; 2724-2729. Sambrook, J.; Fritsch, E.F., Maniates, T. Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory. 1989. Cold Spring Harbor, N Y . Schmidt, F.; Schmid, M . ; Jungblut, P.R.; Mattow, J.; Facius, A . ; Pleissner, K.P. Iterative data analysis is the key for exhaustive analysis of peptide mass fingerprints from proteins separated by two-dimensional electrophoresis. J Am Soc Mass Spectrom. 2003. 14; 943-956. Schreiner, M . ; Strupat, K. ; Lottspeich, F.; Eckerskom, C. Ultraviolet matrix assisted laser desorption ionization-mass spectrometry of electroblotted proteins. Electrophoresis. 1996. 17; 954-61. Schuhmacher, M . ; Glocker, M.O.; Wunderlin, M . ; Przybylski, M . Direct isolation of proteins from sodium dodecyl sulfate-polyacrylamide gel electrophoresis and analysis by electrospray-ionization mass spectrometry. Electrophoresis. 1996. 17; 848-854. Seo, J.; Lee,. K.J . Post-translational modifications and their biological functions: proteomic analysis and systematic approaches. J Biochem Mol Biol. 2004. 37; 35-44. Seymour, J.L.; Lazarus, R.A. Native gel activity stain and preparative electrophoresis method for the detection and purification of pyridine nucleotide-linked dehydrogenases. Anal. Biochem. 1989. 178; 243-247. Shevchenko, A. ; Shevchenko , A . Evaluation of the efficiency of in-gel digestion of proteins by peptide isotopic labeling and M A L D I mass spectrometry. Anal Biochem. 2001, 296; 279-83. Shevchenko, A . ; Wilm, M . ; Vorm, O.; Mann, M . Mass spectrometric sequencing proteins silver-stained polyacrylamide gels. Anal. Chem. 1996. 68; 850-585. Shevchenko, A. ; Wilm, M . ; Vorm, O.; Mann, M . Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal Chem. 1996. 68; 850-8. Soskic, V . ; Gorlach, M . ; Poznanovic, S.; Boehmer, F.D.; Godovac-Zimmermann, J. Functional proteomics analysis of signal transduction pathways of the platelet-derived growth factor beta receptor. Biochemistry. 1999. 38; 1757-64. 104 Stubbs, A.J . ; Skinner, S.L. Lectin chromatography of extrarenal renin protein in human plasma and tissues: potential endocrine function via the renin receptor. J Renin Angiotensin Aldosterone Syst. 2004. 5; 189-96. Sultan, N.A. ; Kenoth, R.; Swamy, M.J. Purification, physicochemical characterization, saccharide specificity, and chemical modification of a Gal/GalNAc specific lectin from the seeds of Trichosanthes dioica. Arch Biochem Biophys. 2004. 432; 212-21. Sundblad, G.; Kajiji, S.; Quaranta, V . ; Freeze, H.H.; Varki, A . Sulfated N-linked oligosaccharides in mammalian cells. III. Characterization of a pancreatic carcinoma cell surface glycoprotein with N - and O-sulfate esters on asparagine-linked glycans. J Biol Chem. 1988. 263; 8897-903. Tanaka, K. ; Waki, H. ; Ido, Y . ; Akita, S.; Yoshida, Y . ; Yoshida, T. Protein and Polymer Analyses up to m/z 100,000 by Laser Ionization Time-of-flight Mass Spectrometry. Rapid Commun Mass Spectrom. 1988. 8; 151-3. Thiede, B.; Hohenwarter, W.; Krah, A. ; Mattow, J.; Schmid, M . ; Schmidt, F.; Jungblut PR. Peptide mass fingerprinting. Methods. 2005. 35; 237-47. Traub, F.; Jost, M . ; Hess, R.; Schorn, K. ; Menzel, C ; Budde, P.; Schulz-Knappe, P.; Lamping, N . ; Pich, A . ; Kreipe, H. ; Tammen, H . Peptidomic analysis of breast cancer reveals a putative surrogate marker for estrogen receptor-negative carcinomas. Lab Invest. 2006. 86; 246-53. Tyers, M . ; Jorgensen, P. Proteolysis and the cell cycle: with this RING I do thee destroy. Curr Opin Genet Dev. 2000. 10; 54-64. UniprotKB/Swiss-prot entry P02662, U R L Vasilescu, J.; Guo, X . ; Kast, J. Identification of protein-protein interactions using in vivo cross-linking and mass spectrometry. Proteomics. 2004. 4; 3845-3854. Vestling, M . M . ; Fenselau, C. Polyvinylidene difluoride (PVDF): an interface for gel electrophoresis and matrix-assisted laser desorption/ionization mass spectrometry. Biochem Soc Trans. 1994. 22; 547-551. Von Seggern, C.E.; Zarek, P.E.; Cotter, R.J. Fragmentation of sialylated carbohydrates using infrared atmospheric pressure M A L D I ion trap mass spectrometry from cation-doped liquid matrixes. Anal Chem. 2003. 75; 6523-30. Wang, J .M.; Ko, C.Y. ; Chen, L.C. ; Wang, W.L.; Chang, W.C. Functional role of NF-IL6beta and its sumoylation and acetylation modifications in promoter activation of cyclooxygenase 2 gene. Nucleic Acids Res. 2006. 34; 217-31. Wang, L . ; Whang, N . ; Wuerffel, R.; Kenter, A . L . AID-dependent histone acetylation is detected in immunoglobulin S regions. J Exp Med. 2006. 203; 215-26. 105 Warren, M.R.; Parker, C.E.; Mocanu, V. ; Klapper, D.; Borchers, C H . Electrospray ionization tandem mass spectrometry of model peptides reveals diagnostic fragment ions for protein ubiquitination. Rapid Commun Mass Spectrom. 2005. 19; 429-37. 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; 242-7. Wilmarth, P.A.; Taube, J.R.; Riviere, M.A. ; Duncan, M . K . ; David, L . L . Proteomic and sequence analysis of chicken lens crystallins reveals alternate splicing and translational forms of beta B2 and beta A2 crystallins. Invest Ophthalmol Vis Sci. 2004. 45; 2705-2715. • Witke, W.; Podtelejnikov, A . V . ; Di Nardo, A. ; Sutherland, J.D.; Gurniak, C.B.; Dotti, C ; Mann, M . In mouse brain profilin I and profilin II associate with regulators of the endocytic pathway and actin assembly. EMBO J. 1998. 1-7; 967-976. Wunsch, A. ; Jackson, V . Histone release during transcription: acetylation stabilizes the interaction of the H2A-H2B dimer with the H3-H4 tetramer in nucleosomes that are on highly positively coiled D N A . Biochemistry. 2005. 44; 16351-64. Yang, Z.; Hancock, W.S. Approach to the comprehensive analysis of glycoproteins isolated from human serum using a multi-lectin affinity column. J Chromatogr A. 2004. 1053; 79-88. Yefimov, S.; Sjomeling, C ; Yergey, A . L . ; L i , T.; Chrambach, A . Recovery of sodium dodecyl sulfate-proteins from gel electrophoretic bands in a single electroelution step for mass spectrometric analysis. Anal. Biochem. 2000. 284; 288-295.. Yip, T.T.; Tarn, Y . Y . ; Kong, Y . C ; Belew, M . C ; Porath, J. Isolation of the epidermal growth factor from the shrew submaxillary gland. FEBS Lett. 1985. 187; 345-8. Young, R.W.; Fulhorst, H.W. Recovery of S35 radioactivity from protein-bearing polyacrylamide gels. Anal. Biochem. 1965. 11; 389-391. Yurewicz, E . C ; Pack, B.A. ; Sacco, A . G . Isolation, composition, and biological activity of sugar chains of porcine oocyte zona pellucida 55K glycoproteins. Mol Reprod Dev. 1991.30; 126-34. 106 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items