2-d proteome analysis protocols

1 2-D Protein Gel Electrophoresis An Overview Jenny Fichmann and Reiner Westermeier Importance Two-dimensional electrophoresis (2-D) of proteins used to be an art practiced by a few researchers, and their worldwide meetings could be held in a side room of a medium-sized hotel With the rapidly growing volume of sequence data produced by the genome projects and the development of new mass spectrometry methods, high-resolution protein analysis has become an important tool in molecular biology High-resolution 2-D can reveal virtually all proteins present m a cell or tissue at any given time, mcludmg those with posttranslational modifications and rapid turnover rates With the new analytrcal tools and the genomic and protein database networks, large-scale studies can be performed on the actual gene products or proteins, in then precursor, mature, and modified forms This task has lately been called the proteome project The proteome projects are the necessary complement to genome analysts, and aim to identify and characterize all proteins expressed by an organism or a tissue (1) 1.1 Definition The term 2-D protein gel electrophoresis is used in this volume primarily to mean the 2-D electrophorests technique in which first-dimension isoelectric focusing in a polyacrylamide gel with a pH gradient and a high concentration of urea (Fig 1) is combmed with a second-dtmension separation on SDS polyacrylamide gels (Fig 2) In the first dimension, the proteins are separated according to their charges (Chapters 14-24a), and m the second dimension, according to their molecular masses(Chapters 25 and 26) The resulting spot From Methods m Molecular Edlted by Biology, A J Lmk Vol 112 2-D Proteome Humana Press Analys/s Inc , Totowa, NJ Protocols Fichmann and Westermeier Fig A schematic showing the isoelectric focusing gel used for the first dimension of 2-D gel electrophoresis Proteins are being focused to their isoelectric point from time b to 12using an immobilized pH gradient patterns are usually oriented according to the Cartesian convention with the low, acidic isoelectric points to the left and the low molecular weights at the bottom (Fig 2) Depending on the 2-D application, different gel formats (Chapters 15, 17, 18,2 I-24a), reducing or nonreducing conditions (Chapter 27 and 28), different pH ranges (Chapters 16 and 22), and different detection methods can be used (Chapters l-38) 1.2 What Does the 2-D Separation Method Offer? Gel electrophoresis has some advantages over other separation techniques Starting materials, such as cell lysates or tissue extracts, can be applied to gels directly and fractionated with very high resolution Electrophoretic techniques exhibit minimal loss of hydrophobic protein species.The separated proteins are embedded in the matrix, where they can be detected with very high sensitivity (essentially unlimited exposure time for fluorography, autoradiography, or storage phosphor imager) The isolated proteins can be readily extracted from the matrix for further characterization by sequence analysis or mass spectrometry (Chapters 48-55) 2-D Protein Gel Electrophoresis t t Fig 2.A schematic showingthe SDS-PAGE usedfor the second gel dimension 2-D gel of electrophoresis equilibration, firstdiiension gel is laid on topof aSDS-PAGE After the gel at time (t,,) and a voltage is applied so that proteins migratesfrom the 1-D gel into the 2-D gel (t,) The 2-D gel is run to separate proteinsrelative to their molecularmass(tz) The two separation parameters of 2-D protein gel electrophoresis (isoelectric point and molecular weight) are essentially orthogonal and independent, even though they occur in the same matrix Either dimension is capable of resolving 100-200 protein species, but the resolving power of the combined techniques is approximately the product of the resolving power of the individual techniques Up to 10,000 individual protein species have been resolved in a single gel (2), similar in magnitude to the estimated number of expressed proteins in a eukaryotic cell (3) or bacterium (4) For studies of minute changes in protein expression or modification in a cell or tissue, it is crucial that the entire array of proteins can be displayed in one gel 1.3 Technical Aspects 2-D electrophoresis usually proceeds in the following order: perform 1-D isoelectric focusing; exchange the buffer in the focusing gel for the SDS buffer; Fichmann and Westermeier place the first-dimension gel m direct contact with the second-dimension SDS gel; perform SDS electrophoresis; detect the separated protein spots The prmciple of an isoelectric focusing separation followed by polyacrylamide electrophoresis was first published in 1969 (5) However, a truly successful 2-D method required the development of an effective sample preparation procedure by Klose (6), and O’Farrell (7) in 1975 Because small variations in various steps of this multistep procedure have a major influence on the resultmg pattern, mechanization and standardization are essential for reproducible results The ISO-DALT system for multiple gel castmg and runmng developed by Anderson and Anderson (89) improved the reproducibility of the technique by addressing a number of mechanical problems in gel casting, loading, and running Another crucial problem with the earlier versions of the 2-D method was the unsattsfactory performance of the first dimension The required pH gradient was established by the mtgratton of individual species from complex mixtures of carrier ampholytes to their respective isoelectrtc points Batch-to-batch variations of the carrier ampholytes resulted in variations m the shapeof the pH gradients Differences in protem and salt concentrations of the starting material influenced the gradient profile During focusing, most of the basic gradient is lost m the buffer reservoirs To visualrze the basic proteins, a specialized 1-D gel run under nonequilibrium conditions is required Many of theseproblems are eliminated with the use of immobilized pH gradients (IPG) (JO) A pH gradient formed by mixtures of acrylamido buffers is covalently fixed to the acrylamtde matrix during gel polymertzation.The gradient does not drift and cannotbe distorted(Chapters19-24a) With this improved first dimension, mtroduced by G&g et al (II), a substantiallywider specttumof proteins can be resolved throughout the entire pH gradient in one gel The improved stability and reproducibUy of the gradient and commerctal availabihty of precast gradient gels allowed comparisons to be made between gels run in the same lab at different times as well as to gels from outside laboratories run under the same conditions (12) For the second dimension, SDS gels of different compositions can be used depending on the sizeof the proteins of interest For most applications, the discontinuous Tris-HCYTris-glycme buffer systemof Laemmh IS employed Variations include substituting acetate for HCl to improve gel shelf-life with a pH value below neutrality, and substituting tricine for glycine to improve low-molecularweight pepttde separations The resolution of the SDS dimension can be further optimized for particular molecular weight ranges by introducing a stacking gel and employing acrylamide concentration gradient gels (Chapters 25 and 26) 1.4 Sample Preparation Adequate sample treatment is the most important prerequisite for a successful 2-D experiment The sample preparation procedure must* 2-D Protein Gel Electrophoresis 3, Stably solubilize all proteins, includtng hydrophobic spectes Prevent protein aggregation and hydrophobic interactions Remove or thoroughly digest any RNA or DNA Prevent artifactual oxidatton, carbamylatron, proteolytic degradatron, or conformational alteration Each naturally occurring polypeptide should be represented by only one spot in the gel In general, a cell lysis (or sample solubihzation) buffer contains 8-9.8 M urea, a nonionic or zwitterionic detergent, carrier ampholytes, dithiothreitol, and, depending on the sample, protease inhibitors and/or protease-free nucleases There is no universal protocol for sample preparation Different sample sources requrre different extraction and lysis techniques (Chapters 2-l 1) For autoradiography/fluorography detectron, cell proteins must be labeled with radioactive isotopes through growth in the presence of the appropriately radiolabeled precursors Sample protein concentrations are usually determined before the first dimension (Chapters 12 and 13) 7.5 Protein Load The amount of proteins applied to a gel can vary between several mrcro- grams to g If a minor component must be detected against a background of abundant proteins, such as albumin, m a serum sample, a high-protein-capacity system is required Capacity IS dependent on the volume of the gel Thinner gels provide better sensitivity for most detection methods, and larger and thicker gels offer increased capacity Whether the sample should be loaded on the anode or cathode end of the isoelectric focusing gel must be determined experimentally for each new sample An interesting new approach combines rehydration of a precast dry immobilized pH gradient strip with sample application The protein sample IS mixed with the rehydration buffer and the IPG strip rehydrated m the mixture including the sample The proteins are distributed over the entire pH gradient Regardless of where proteins start in the pH gradient, they migrate in the electric field to their corresponding isoelectric points (Chapters 24 and 24a) This approach works particularly well when semipreparative and preparative amounts of sample must be loaded (13) 1.6 Instrumentation The classical setup for the 2-D technique according to O’Farrell uses vertical thin gel rods for the first dimension and vertical slab gels for the second dimension With the ISO-DALT system (8), up to 20 gels can be cast and run in parallel Isoelectric focusing on the immobilized pH gradient gel strtps on film supports is more convenient in a horizontal format The IPG gel strips can be Fichmann and Westermeier used with either vertical or horizontal second-dtmension SDS gels Which system should be chosen? Vertical slabs are superior for high protein loads and multiple runs Horizontal systems employ film-supported gels, which not change their dimensions during staining and drymg, and can be very thm for high resolution and sensitivity of detection Proteins m very thin gels (300 kDa (Chapters 50-52) Other widely used identification techniques include comigration of the protein to be identified (Chapter 45); immunoblottmg (Chapter 36 and 46); amino acid analysis (Chapter 47); N-terminal sequencmg (Chapter 48); and pepttde fingerprmting by partial in-gel or eluate digestion followed by SDS-PAGE or reversed-phaseHPLC separation (Chapter 45 and 49) 2-D Protein Gel Electrophoresis References Wilkins, M R , Sanchez, J C., Gooley, A A., Appel, R D , Humphrey-Smith, E., Hochstrasser, D S., and Williams, K L (1995) Progress with proteome proJects why all proteins expressed by a genome should be identified and how to it Blotechnol Gene Eng Rev 13, 19-50 Klose, J and Kobalz, U (1995) Two-dimensional electrophoresis of proteins an updated protocol and tmpltcations for functtonal analysts of the genome Electrophoreszs 16, 1034-1059 Cells, J E , Rasmussen, H H , Gromov, P., Olsen, E., Madsen, P., Leffers, H , Honor, B , Dejgaard, K., Vorum, H , Knstensen, D B , Ostergaard, M , Haunse, A., Jensen, N A., Cells, A., Basse, B., Lauridsen, J B , Ratz, G P., Anderson, A H., Walbum, E., Kjaergaard, I , Andersen, I , Puype, M , Van Damme, J., and Vanderkerckhove, J (1995) The human keratinocyte two-dimensional gel protein database (update 1995) Mapping components of signal transduction pathways Electrophoreszs 16,2 177-2240 Pasqualt, C., Fruttger, S , Wilkms, M R , Hughes, G J , Appel, R D , Batroch, A , Schaller, D., Sanchez, J -C , and Hochstrasser, D F (1996) Two-dtmenstonal gel electrophoresis of Escherzchza coli homogenates the Escherchza colz SWISS2DPAGE database Electrophoresis 17, 547-555 Macko, V and Stegemann, H (1969) Mappmg of potato proteins by combined electrofocusmg and electrophoresis Identification of varieties Hoppe-Seyleris Z Physzol Chem 350,9 17-9 19 Klose, J (1975) Protein mappmg by combined tsoelectrtc focusing and electrophoresis of mouse tissues Humangenetik 26,23 l-243 O’Farrell, P H (1975) High resolution two-dimensional electrophoresis of proteins J Bzol Chem 250,4007-4021 Anderson, N G and Anderson, N L (1978) Analytical techniques for cell fractions XXI Two-dimensional analysts of serum and tissue proteins Multiple isoelectric focusing Anal Biochem 85, 33 l-340 Anderson, N L and Anderson, N G (1978) Analytical techniques for cell Fractions XXII Two-dimensional analysts of serum and ttssue proteins Multiple gradient-slab gel electrophoresis Anal Bzochem 85, 34 l-354 10 Bjellqvist, B , Ek, K., Righetti, P G., Gianazza, E., Gorg, A , Westermeier R , et al (1982) Isoelectrtc focusmg m immobtlized pH gradients principle, methodology and some applications J Bzochem Bzophys Methods 6,3 17-339 11 Gorg, A., Postel, W , and Gunther, S (1988) The current state of two-dimensional electrophoresis with immobilized pH gradients Electrophoreszs 9, 53 l-546 12 Blomberg, A., Blomberg, L , Norbeck, J., Fey, S J , Larsen, P M , Roepstorff, P., Degand, H., Boutry, M., Posch, A., and G&g, A (1995) Interlaboratory reproducibility of yeast protein patterns analysed by immobilized pH gradient two-dimenstonal gel electrophorests Electrophoreszs 16, 1935-1945 13 Rabilloud, T , Valette, C., and Lawrence, J J (1994) Sample apphcatton by mgel rehydration improves the resolution of two-dimensional electrophoresis with nnmobtlized pH gradients m the first dimension Electrophoreszs 15, 1552-1558 Solubilization of Proteins in 2-D Electrophoresis An Outline Thierry Rabilloud Introduction The solubilization parallel goals: process for 2-D electrophoresis has to achieve four Breaking macromolecular interactions in order to yield separate polypeptide chains: This includes denaturing the proteins to break noncovalent interactions, breaking disulfide bonds, and disrupting noncovalent interactions between proteins and nonproteinaceous compounds, such as lipids or nucleic acids Preventing any artifactual modification of the polypeptides m the solubihzation medium: Ideally, the perfect solubihzation medium should freeze all the extracted polypepttdes m their exact state prior to solubtlization, both in terms of ammo acid composition and m terms of posttranslational modifications This means that all the enzymes able to modify the proteins must be quickly and irreversibly inactivated Such enzymes include of course proteases, which are the most difficult to mactivate, but also phosphatases, glycosidases, and so forth In parallel, the solubilizatton protocol should not expose the polypepttdes to condmons m which chemical moditications (e.g., deamidation of Asn and Gln, cleavage of Asp-Pro bonds) may occur Allowing the easy removal of substances that may interfere with 2-D electrophoresis In 2-D, proteins are the analytes Thus, anything m the cell but proteins can be considered an interfering substance Some cellular compounds (e.g., coenzymes, hormones) are so dilute they go unnoticed Other compounds (e g , simple nonreducing sugars) not interact with proteins or not interfere with the electrophoretic process However, many compounds bind to proteins and/or interfere with 2-D, and must be eliminated prior to electrophoresis if their amount exceeds a critical interference threshold Such compounds mainly include salts, lipids, polysaccharides (mcludmg cell walls), and nucleic acids From Methods m Molecular Bdogy, Vol 112 2-D Proteome Analysrs Edlted by A J Lmk Humana Press Inc , Totowa, NJ Protocols Rabilloud Keeping proteins in solution during the 2-D electrophoresis process Although solubilization strzcto sensu stops at the pomt where the sample is loaded onto the first-dimension gel, its scope can be extended to the 2-D process, per se, since proteins must be kept soluble until the end of the second dimension Generally speaking, the second dimension is an SDS gel, and very few problems are encountered once the protems have entered the SDS-PAGE gel The one main problem 1s overloadmg of the major proteins when mtcropreparative 2-D IS carried out, and nothing but scalmg up the SDS gel (its thickness and its other dimensions) can counteract overloading an SDS gel However, severe problems can be encountered m the IEF step They arise from the fact that IEF must be carried out m low tonic strength condmons and wtth no mampulatron of the polypeptide charge IEF condttions give problems at three stages a During the mittal solubthzation of the sample, tmportant mteracttons between proteins of widely different pls and/or between protems and interfering compounds (e.g., nucleic acids) may happen Thts yields poor solubrllzatton of some components b During the entry of the sample m the focusmg gel, there is a stackmg effect owing to the transition between a liquid phase and a gel phase with a higher friction coeffictent Thts stackmg increases the concentration of protems and may give rise to precipitation events c At or very close to the tsoelectric pomt, the solubihty of the proteins comes to a mmtmum This can be explamed by the fact that the net charge comes close to zero, with a concomttant reduction of the electrostatic repulsion between polypepttdes This can also result m protein precipitation or adsorption to the IEF matrix Apart from breaking molecular interactions and solubtlrty in the 2-D gel, which are common to all samples,the solubilization problems encountered ~111 greatly vary from one sample type to another owing to wide differences m the amount and nature of interfering substances and/or spurious actrvmes (e.g., proteases) The aim of this outline chapter is not to give detailed protocols for various sample types, and the reader should refer to the chapters of thts book dedicated to the type of sample of interest, The author would rather like to concentrate on the solubrlization rationale and to describe nonstandard approaches to solubilization problems A more detailed review on solubthzation of proteins for electrophoretic analyses can be found elsewhere (I) Rationale of Solubilization-Breaking Molecular Interactions Apart from disulfide bridges, the main forces holding proteins together and allowmg binding to other compounds are noncovalent mteracttons Covalent bonds are encountered mainly between proteins and some coenzymes The noncovalent interactions are mainly ionic bonds, hydrogen bonds, and “hydrophobic mteracttons.” The basis for “hydrophobtc interactions” is in fact the Jensen et al 574 3.2 A Sfep-by-Step Guide fur Installing the Nanoelecfrospray Source Mount the pulled and metal-coated nanoelectrospray needle m the holder Connect the holder to the ton source power supply Make electrical contact between the needle and the holder by applying a droplet of graphite paste (Alternattvely, the power supply can be connected directly to the needle with a clamp.) Mount the ton source on a x-y-z mampulator m front of the mass spectrometer Dtssolve proteins and peptldes m 5% fornnc acid m 20-50% methanol, and inject S-2 pL mto the nanoelectrospray needle using a mlcroptpet with a gel loader ttp (see Note 3) Ltqutd inJected into the needle is drawn to the tip by capillary force Connect the needle holder to the 20-mL syringe, which provides the air pressure Posttion the nanoelectrospray needle on axis and 5-2 mm from the orifice of the mass spectrometer Momtor the position of the needle tip with a mtcroscope or a video camera Gently pressurize the needle by an using the 20-mL syringe If no liquid appear at the needle ttp, then briefly and gently touch it against the Interface plate of the mass spectrometer under macroscopic control The needle should not be visibly shortened, but a small sample droplet may appear after the contact indicating the opemng of the needle ttp (see Fig and Note 4) Reposition the needle m front of the ortfice of the mass spectrometer 10 Apply the voltage to needle/holder and mass spectrometer interface, and start scanning the mass spectrometer (see Note 5) If there are ions rn the spectrum, reduce the air pressure m the needle to the lowest value that still keeps the flow stable If no tons appear, then refer to Note for troubleshootmg tips 3.3 Sample Preparafion for Nanoelecfrospray Mass Specfromefry The nanoelectrospray ton source unfolds tts full potential when the avarlable sample IS concentrated to pL or less Centrifugal mtcrocolumns can be used to desalt and concentrate protein and peptrde samples to & volumes (6,7) A pulled glass capillary IS packed with a few hundred nanoliters of Poros resin Working m the perfusion mode, Poros material generates only a small flow resistance when packed into a capillary Peptide soluttons are normally desalted/concentrated on Poros R2 resin, protein solutions on Poros Rl resin, and hydrophilic samples and small pepttdes on OhgoR3 material The pulled glass captllarles used for the columns are the same as used for nanoelectrospray emitters, but they are not sputter-coated with metal For peptide analysis, Poros R2 material 1sused as chromatographlc resin Remove the smallest particles by sedimentmg the resin three to five times m methanol In a 1.5~mL microcentrtfuge tube, make a slurry of IO-20 $ resin in 1.2 mL methanol Mount a pulled glass capillary into a custom-made capillary holder (Fig, 3) or into a pierced lid of a 1.5-mL mtcrocentrtfuge tube Use a micropipet with a gel loader tip to transfer & of resin slurry mto the capillary Peptide Sequencing 575 Capillary, loaded with chromatographic resin Nano-Electrospray lmm lcm Fig Centrifugal capillary column and transfer assembly The capillary column used for desalting/concentration of sample is made from a pulled glass capillary and tilled with a small volume of Poros resin The sample is rinsed and then eluted from the column capillary directly into the nanoelectrospray needle by gentle centrifugal force Load the chromatographic resin into the tip of the glass capillary by centrifugal force using a manually operated tabletop minicentrifuge at low speed (500-2000 rpm) The chromatographic material is visible against a dark background When sufficient chromatographic resin has been loaded into the capillary (l-2 mm resin height), the glass tip is widened/broken by gently touching it against the tabletop The opening should allow liquid but not column resin to exit the capillary during centrifugation Do not centrifuge the resin too fast Otherwise it is compressed and may block the flow of liquid The centrifugal capillary column is only used once to avoid sample-to-sample contamination Rinse the capillary column by injecting pL 50% MeOH followed by gentle centrifugation Equilibrate the column resin by injecting pL 5% formic acid, into the capillary followed by centrifugation until all liquid has passed through the column Dissolve the sample in 10-20 pL 5% formic acid, and inject it onto the capillary column in aliquots of pL followed by centrifugation Wash the column resin twice by centrifugation with pL 5% formic acid solution This desalting step is very efficient, since the column is washed with 50-100 times its resin volume Before beginning the elution step, the washing solution must be completely removed by gentle centrifugation Mount the capillary co1umn in-line with a premade nanoelectrospray needle in a custom-made capillary holder that tit into a microcentrifuge (Fig 3) Jensen et al 576 10 Elute the peptlde mixture into the nanoelectrospray capillary by centrifuging twice with 0.5 pL 60% methanol/5% formic acid Elute proteins with 60-70% acetomtnle/5% formic acid This procedure allows handling of elution volumes between 10 and 0.2 pL+ Elution, however, should be performed twice, because the first elutlon does not completely deplete the column Keep m mmd that signal Intensity m an electrospray spectrum is concentration dependent so keep the elutlon volume as small as possible 11 Mount the loaded nanoelectrospray needle onto the ion source and begin the experiment (see Note 9) 3.4 Nanoelectrospray Tandem Mass Specfrometry of Unseparated Peptide Mixtures Peptide sequencing with tandem mass spectrometry (15,16) consists of three steps: Measuring the m/z values of peptldes in a sample Acquiring the tandem mass spectra after colllslon-induced mentation) of selected peptides Interpreting the tandem mass spectrometry data dlssoclation (1.e , frag- With the nanoelectrospray source, the first two steps are performed in one experiment with the unseparated peptlde mixture The m/z values of analyte peptides are detected by comparing a Q, mass spectrum to a representative spectrum of the autolytlc peptldes of the enzyme used (i.e., trypsm) or a representative control from a particular experiment It is often advantageous to process an empty gel piece excised near the protein band of interest as control (see Chapter 52) At subplcomole protein amounts on the gel, we routmely employed the parent ion scan techmque to detect peptide ion signals that were below the chemlcal noise in the normal Q, spectrum (Fig 4) (I 7) Parent ion (or precursor ion) scans of the abundant m/z 86 immonium ion of lsoleucme/leucine detect peptides that contain these common amino acids For the parent ion scan, the mass spectrometer parameters are adjusted to obtain optimum detection efficiency at reduced mass resolution The parent ion scan technique can also be used to detect selectively phosphopeptldes by monitoring m/z 79 in the negative ion mode (17-19), to detect selectively glycopeptldes by monitoring m/z 162 or 204 m the positive ion mode (17,20), and to detect intact proteins or ollgonucleotides in contaminated samples (21) Once a set of peptide m/z values has been determined by either Q, scans or by parent ion scans,htgh-resolution scanscan be performed for selected peptide ion signals in the “multiple ion monitoring” mode to determine the exact peptide mass and the peptide charge state based on the isotope spacing The reduction of sensitivity when measuring at high resolution IS compensated by adding many scans, for example, 50 or more, to one spectrum The latter fea- Peptide Sequencing 577 A Polymer Contamination 100 iz d &j 50 460 B 600 650 600 660 700 650 700 Peptides from the Protein 100 L% c 22 60 450 500 560 600 Fig Parent ion scan for the leucme/isoleucine immomum ion (m/z 86) detects peptlde ions below the chemical noise level (A) The nanoelectrospray mass spectrum (Q, scan) of a tryptic peptide mtxture displays only polymer ion signals and chemical background, which suppress peptide ions (B) The parent ion (m/z 86) analysts of the same sample reveals several peptide ion stgnals, which subsequently can be selected for sequencing by tandem mass spectrometry ture further demonstrates the utility of long measurement times that the nanoelectrospray source provides, It is advantageous to select doubly charged tryptic peptide ions for tandem mass spectrometry experiments, because they generate relatively simple fragment ion spectra Trrply charged tryptic peptides can also be fragmented and often allow determination of long stretches of amino acrds sequence, i.e., 15-25 consecutive residues, vra doubly charged fragment ion series 3.5 Fragmenting Peptides by Collision-Induced Dissociation Once a set of peptide m/z values has been accurately determined, each peptide IS fragmented in turn For peptide sequencing by tandem mass spectrometry using a triple quadrupole mstrument, two main instrumental parameters are adjusted to obtain high-quahty amino acid sequence information Fn-st, the resolutton setting of the first quadrupole (Qr) is adjusted according to the abundance of the pepttde ion signal, i.e., the lower the ton intensity, the higher the 578 Jensen et al r 72 100 eV 7I 200 55.2 eV )I 300 400 500 44.3 eV -I 600 700 800 900 ! 1000 1100 Massto Charge m/z Fig Peptidetandemmassspectrumacquiredin separate segmentsThe m/z range abovethe precursorion, [M + ZH]‘+, is acquiredat low colhslon energyandat relatively low massresolution to generateanddetectlarge peptide fragment ions efficiently The m/z range below the precursor ion IS acquired at higher resolutron and with stepped collision energies,1e., intermediatecollision energyto generatelow m/z sequence ions andhigh collision energyto generatennmomumions andthe a2and b2fragments resolution setting in order to reduce the chemical background noise in the lower half of the tandem mass spectrum for a better signal-to-noise ratio Second, the collision energy can be adjusted according to the peptide mass and varied depending on the mass range scanned (Fig 5) The collision gas pressure is kept constant throughout the MS/MS experiment It may be advantageous to acquire a tandem mass spectrum m two or three segments, The high m/z segment IS acquired with a wide parent ion selection window (low resolution) and a low collision energy to generate and detect relatively large peptide ton fragments The low m/z region is acquired at higher resolution and at higher colhsion energies to generate and detect low m/z fragments and ammonium ions The nanoelectrospray allows this and other types of optimization owing to the stability and long duration of the spray When mvestlgatmg a peptrde mixture by tandem mass spectrometry, as many peptrdes as possible should be fragmented Thus motivated the development of semtautomatic software routines to assist in data acqursltion The list of peptide m/z values is stored by customized software that calculates the optrmum hardware settings for subsequent sequencing of each individual peptide However, for de y1ovosequencing of long stretches of ammo acid sequences, it is not yet advisable to use automated software routines for data acqursition Careful adjustments of colhsion energy and mass resolution is required to obtain high-quality data for unambiguous sequenceassignments Only when very high data quality is obtained, e.g., when employmg a quadrupole time-of-flight tandem mass spectrometer m combmation with 180-labeling (22), are software routines reliable Peptide Sequencing 579 (M+ 2H)2+ 100 939.0 50 i F2 OII,* +-I* 200 100 591.4 300 400 500 E F 600 700 I.**800-L -A 900 CC” -‘ ’ 1000 1100 Mass to Charge m/z ml partial sequence ,,zL .I Sequence in database Fig Peptidesequence generatedfrom a tandemmassspectrum.A short search tag string, (591.4)FEA(939.0), is readily identified in the tandem massspectrum of this tryptic peptide with massM It is subsequently converted to a peptide sequencetag by the PeptideSearch software andusedto query a database modular composition of The a peptide sequencetag permits error-tolerant searches where one or two of the modules are allowed to contain an error 3.6 Generation of Peptide Sequence Tags from Tandem Mass Spectra of Peptides Complete interpretation of tandem mass spectra of peptides can be complicated and requires some experience However, it is often relatively straightforward to generate short sequences of two to five amino acid residues from a tandem mass spectrum This information is valuable for sequence database searches as follows A “peptide sequence tag” is assembled from the peptide mass, a short internal sequence of amino acid residues, and the distance in mass to the N- and C-terminus of the peptide (23) (Fig 6) The search specificity of this construct is very high, because the amino acid sequence is “locked” in place by the masses of the “unknown” parts of the peptide The modular composition of a peptide sequence tag makes it tolerant to errors in any one of the modules Since only a fraction of the information content of the tandem mass spectrum is used to generate a peptide 580 Jensen etal sequence tag used to query the sequence database, the remaining mformatron confirm a retrieved peptide sequence: Every significant fragment ion signal should correlate to the peptide sequence A peptrde sequence tag consisting of three residues typically retrieve only one protein from a database containing more than 200,000 sequences If longer stretches of sequence can be read out of a tandem mass spectrum, i.e., six or more residues, tt is advantageous to search by ammo acrd sequence instead of by peptrde sequence tags Searching by ammo acid sequence 1s more flexible and allows sequence homology searches As mentioned above, tryptrc pepttdes have the desirable feature that they contam an N-termmal ammo group and a C-terminal Lys or Arg residue, localizing protons at both the N-termmus and the C-termmus of the pepttde Tandem mass spectra of tryptlc pepttdes very often contam a contmuous y-ion series, which can be readily assigned in the m/z range above the doubly charged parent ton signal Our spectrum interpretation strategy builds on this characteristic It IS guided by the demand to identify a protein m sequence databases or to sequence peptides reliably for cloning of the cognate protein The following list summartzes a few basic empirical rules that we use in mterpreting tandem mass spectra of tryptic peptides Since peptrdes differ in then fragmentation behavior m a sequence-dependent manner, it is possible to find exceptions to these rules The goal of the interpretation is to find a series of peaks that belong to one ion series-for tryptic pepttdes mostly y-ions (C-termmal fragments) Initial peak selection: The high m/z region of a tandem mass spectrum IS often straightforward to interpret Choose an intense ion signal in this regron as the “starting peak ” Assembly of a partial amino acid sequence* Try to find ion signals that are precisely one amino acid residue mass away from the starting peak (up or down m mass) We use software that marks all the possibrhties m the spectrum This provides a good overview whether there IS more than one posstbiltty for sequence assignments If there is a repeating pattern of lidgment ion peaks with satellite peaks representing an Hz0 loss (-18 Dalton) or an NH, loss (-17 Dalton), a fragment ton series has been identified (for tryptlc pepttdes a y-ion series is more likely) By repeating step 3, a peptide sequence tag consisting of two to four ammo acids is assembled, which is subsequently used to identify a protein m the sequence database PeptideSearch software automatically assembles a sequence tag from a series of fragment ion masses (23,24) As default for tryptic peptrdes, the database is searched under the assumption that a y-ion series was determined However, even for tryptic peptides, the tandem mass spectrum can be dominated by b-ions if a pepttde contained an internal basic residue or when the C-terminal peptide of the protein had been sequenced Peptide Sequencmg 581 3.7 Confirming Protein /den tifications Made by Peptide Sequence Tags If a protein sequence is retrieved by a database search with a peptide sequence tag, then the amino acid sequence of the retrieved peptide should fit the tandem mass spectrum in order to be called a positive match Two or more peptides from a sample should independently identify the same protein m a database.To verify a match, the peptide fragment massesmust be correct withm the expected error of the mass measurement For tryptic peptides, the y-ion series should be nearly complete, except when a peptide contains an internal prolme residue (see below) The N-terminal b2 and a2fragment ions, generated at relatively high colhsion energy, should be present m the low m/z region Odd fragmentation patterns should reflect the ammo acid sequence as discussed in the following paragraphs Peptides that contain internal basic residues (lysine or arginme) not fragment in the vicinity of these residues, because a charge is localized at the side chain of the basic residue and therefore not available for amide backbone cleavage If the triply charged precursor ion was fragmented, then doubly charged y-ions are present in the spectrum Internal proline residues deserves special attention Cleavage of the C-terminal bond of a proline is observed to a low degree The N-terminal bond of a prohne is labile, giving rise to an intense y-ion fragment Internal fragmentation of peptides contaming a prolme often confirms a sequence: The y-ion generated by fragmentation at the N-terminal side of Pro will dissociate a second time to produce a y-ion series, which is superimposed on the origmal y-ion series However, the internal immonium tons (b-type ions) generated from this Pro-containing peptide fragment serve to confirm the C-termmal part of a peptide sequence up to and including the internal proline residue Isoleucine and leucine cannot be differentiated by amide bond cleavage alone, because they have the same elemental composition and therefore identical molecular weight Pairs of amino acid with identical nominal mass, Lys/Gln (128 Dalton) and oxidized Met/Phe (147 Dalton) can often be distinguished Lys and Gln differ in their basicity, and trypsin cleaves C-terminal to Lys and not at Gln, so internal Lys is rarely found in tryptic peptides However, if the latter is the case, then the tryptic peptide usually acquires an additional proton for a total of three charges, and the tandem mass spectrum will often contain a doubly charged y-ion series, If an internal Lys residue is suspected, then the peptide mixture can be inspected for the presence of the limit peptide produced by tryptic cleavage at this Lys residue, Oxidized methionme (147.02 Dalton) and phenylalanine (147.07 Dalton) residues are differentiated relatively easily as follows Tandem mass spectraof peptides that contain an oxidized methionine residue (i.e., methionme sulfoxide) display satellite peaks,which appear64 Dalton 582 Jensen et al below each methtonine sulfoxide-contauung y-ton fragment owing to loss of CH3SOH from methtonine sulfoxide (25,26) The oxidatton reaction 1s often not complete, so mspectronof the peptide massspectrum (Qt spectrum) may reveal a peptrde ion signal 16 Dalton below the one containing oxidtzed methtonme If no proteins are retrieved with a simple database search, then search under the assumption that some of the ammo acids of the pepttde are modified, for example, oxidized methionine or S-acrylamtdocysteme An error-tolerant search can be launched m which only partial correspondence between the peptide sequence tag and a database entry 1srequired (23) Additional mformatton about the protein can be used to select posstble candidates tf more than one protein sequence 1sretrieved (such as protein size, organism, or function) If a protein cannot be tdentttied by any of its pepttde sequence tags, we conclude that tt 1sunknown For proteins from human, mouse, or other model organisms, the peptide sequence tags are then screened against a database of expressed sequence tags (ESTs) (27) ESTs are short stretches of cDNA, t.e , single-stranded DNA generated from expressed mRNA by reverse transcrtptase, and thus represents the set of expressed genes in a given cell line If a database search retrteves a cDNA sequence, then library screening and clonmg are relatively straightforward If the EST database search produces no hit, then de novo peptide sequencing has to be pursued (IO) 3.8 De Nova Peptide Sequencing by Nanoelectrospray Tandem Mass Spectrometry To sequence unambiguously an unknown protein for homology searching and clonmg, two data sets are usually required (20,15) The trypttc peptide mixture 1ssplit m two porttons The first portion of the mixture 1sanalyzed by nanoelectrospray tandem mass spectrometry, and long peptide sequences are generated through complete mterpretatton of tandem mass spectra using the guidelines described above The other portion of the pepttde mtxture is U-methylesterified @5,28) and then analyzed Every free carboxyl-group, mcludmg the C-termmus ofpepttdes, is estertfied and therefore increasesm massby 14 Dalton The number of methyl-esters m a peptrde can be determined by the mass shift of pepttdes, which 1spredictable from the previously interpreted tandem mass spectra of the native peptides Because all y-ton fragments produced from an esterified peptlde contain the C-terminus, they are all shifted up in mass Comparison of a set of tandem mass spectra obtained from a peptide and the corresponding esterified pepttde serve to confirm the amino acid sequence, because the y-ion series can unambiguously be assigned (Fig 7) Additionally, internal acidic residues, Asp and Glu, are methylated as well and can easily be dtfferentiated from then corresponding amide residues, Asn and Gin, which otherwise differ m mass by only Dalton 583 Peptide Sequencing Unmodified Peptide K M A P D Y W Y I’ T YJJj; iai;b,,[i;,;i ,,, 600 600 700 a00 900 Esterified Peptide 100 F - Go- P; 26- y’;* ‘) ,* ** Y6 Yq’ IG- Y;* I II ** y7 4,** Y6 ,, *+ Y;* ya Y”‘* , ‘I &.+ 10 20 30 40 60 610 Mass to Charge 70 80 1000 1100 1200 Ratio m/z Fig Tandem massspectra of (A) natrve and (B) O-methylestenfted peptlde Methylatton of peptide carboxyl groups results m a massincreaseof 14 Dalton of all peptlde fragment tons that contain actdrc restdues, mcludmg the C-terminal y-ions This approach identifies y-ion series and resolves AsplAsn and Glu/Gln ambigutties The number of acidic residues rn peptide ton fragments IS mdlcated with asterisks An alternative method to recognize y-ion series employes **O-labeled water (see, for example, 29,30) By performing the trypsin digestion m a 1: mixture of normal water and ‘*O-labeled water, all the tryptic peptrdes will incorporate l*O at the C-termmus with a yield of approx 50% Each peptide ton appears m a mass spectrum as a doublet separated by Dalton Selecting both isotope species together for fragmentation (low-resolution setting m Q, precursor ton selection) produces tandem mass spectra that display y-ions as a series of split peaks, i.e., separated by Dalton, whereas b-ions are single peaks This aids m the mterpretatron of a peptrde tandem mass spectrum, because y-ton series are unambiguously assrgned This approach ISvery attractive when using a hybrid quadrupole-TOF tandem mass spectrometer, which produces highly resolved fragment ion signals (22) Note that the ‘80-labeled water has to be very pure to avoid chemical background noise Redistillation of commercially available I*O-labeled water is recommended 3.9-O-Methyiesterificafion of Peptides This approach works well for methyl esters only Peptide esterificatton by ethyl alcohol is mcomplete 584 Jensen et al 3.9 I Preparation of Reagent Cool mL of distilled (i.e , dry) methanol m a 5-mL mlcrocentrlfuge tube m the freezer (-20 or -8O’C) for 10-15 Dropwise add 100 pL of acetyl chloride Beware that the reaction mixture may react violently and spray liquid around (or on) you Allow the reagent cocktail to warm up to room temperature, and use it 10 mm later 3.9.2 Derivatization Dry the peptlde (or peptlde mixture) in a vacuum centrifuge Add 2-5 pL of the reagent If the sample contains salts, then add enough reagent to cover the solid residue at the bottom of the tube Incubate at room temperature for 30 Dry the sample in a vacuum centrifuge Esterified peptldes can be redissolved in 5% formic acid They are relatively stable under acidic conditions, but can undergo rearrangement at mild alkaline pH Notes The first heating/pulling stage reduces the diameter of the capillary to about 0.5 mm, but the second stage pulls the glass capillary apart, producing two nanoelectrospray needles These needles should have an opening of l-2 pm However, after pulling, the opening diameter can be Cl00 nm and has to be widened (Subheading 3.2.) To reduce the flow resistance for a stable flow rate in the 10-25 nL/mm range, the narrow part of the tip should not be longer than 500 pm (Fig 1) Needles with very short constrictions (50-100 pm) can be operated easily, but with a higher risk of loosing sample owing to a higher flow rate We prefer such short needles for rapid mass measurements when abundant sample 1s available, e.g., recombinant proteins, synthetic peptides, or ohgonucleotldes Longer tips (200-500 w) are used for tandem mass spectrometry experiments when the longest possible operation time 1sdesirable and when the sample load volume ~111 not exceed pL A major advantage of these types of nanoelectrospray needles 1sthat they not easily block owing to the relatively short length of the needle tip Nanoelectrospray needles and ion sources are commercially available from Protana A/S (Odense, Denmark) The needles are only used once, so it is not a problem that the coating is not tightly fixed to the glass and can be rubbed off Methods to produce a more stable metal coating include pretreatment with (3-mercaptopropyl)trlmethoxysllane (31) or protecting the metal layer by a second layer of $0, (32) A stable gold coating is necessary when a glass needle is used for several samples over a prolonged time The needle tip is fragile, so take care not to break it when loading sample and mounting the needle m the holder The electrospray generated with the nanoelectrospray ion source is very stable This allows purely aqueous solutions to be sprayed even in negative ion mode without nebulizer assistance The source can be operated with solutions contam- Peptide Sequencing 585 ing up to MNaCl (61, although this is not recommended The high stabthty allows optimization of experimental conditions based on analyte characteristics rather than electrospray requirements, i.e., when choosmg buffer cornpositron For example, preservation of noncovalent complexes often does not allow addltion of organic solvent to the sample to facilitate spraying Because the ion source exclusively produces very small droplets, relatively soft desolvation conditions in the interface region of the mass spectrometer can be chosen Needle and plate should be at the same electric potential, When the needle tip is opened, a tiny droplet may appear on the metal plate-n spreads out as a famt shadow m a few seconds The mam objective when operating the nanoelectrospray source IS to achieve a low and stable flow rate despite sample-to-sample variations Thus is achieved by applying air pressure to the needle, which helps to adjust the flow rate and thereby compensates for differences in needle orifice size and sample vtscoslty The flow rate of a nanoelectrospray source is about 25 nL/min At this flow rate, analyte concentrations of pmol/pL results in one analyte molecule per droplet on the average (5) Furthermore, pL of sample is consumed in 40 min, extending the ttme available for opttmizatron of experiments The overall sensitivity is limited by the srgnal-to-notse level, and It is therefore a function of the ionization efficiency, desolvation efficiency, ion transmission, the level of chemrcal background ions, and detector characteristics Several parameters were changed in the inlet region of the mass spectrometer in order to operate the nanoelectrospray ion source These changes are described below: a The nanoelectrospray source IS mounted directly in front of the orifice of the mass spectrometer, on axis with the mass analyzer, and at a distance of 1S-2 mm from the orifice Electrospray at a low flow rate generates very small droplets wtth a dtameter of 200 nm or less Desolvatlon is therefore achieved m a very short time and distance b To initiate the electrospray, a mmtmal electrical field strength at the surface of the liquid has to be reached (33) Conventional electrospray ion sources are operated with a 3- to 5-kV potential difference between the needle and counterelectrode (i.e., the orifice plate of the mass spectrometer) The very small tip diameter of the nanoelectrospray needle allows a spray cone to be established at a much lower electrical potential, typically 400-700 V c Because the charged droplets generated by the nanoelectrospray ion source are very small, softer desolvation conditions in the interface (sknnmer) region of the mass spectrometer are used The electrical gradient and the countercurrent gas flow can be reduced Thts appears to lead to the generation of colder molecular ions as compared to conventtonal electrospray sources facilitating, for example, studies of noncovalent molecular interactions The API III triple quadrupole instrument (Perkin-Elmer/Sciex) has a channeltron detector, which is operated in smgle-ion countmg mode At a concentration of usually 10 fmol/pL, the limit for peptide detection in Q, scan mode (m/z 400-1500) with a signal-to-noise ratio larger than is reached Usmg parent or precursor ion 586 Jensen et al scan techmques, the sensmvtty can be extended to fmol/pL Detectton of proteins requrres hrgher analyte concentration, because the ton current is spread over multiple charge states and the molecules are more difficult to desolvate The nanoelectrospray ton source IS a very robust devrce However, problems may occur if the sample contains high concentratrons of buffers, polymers, or salts, or if the shape of the needle tip 1snot wtthm the dtmenstons descrrbed above In this section, we provide a few troubleshootmg tips If there are no tons or no noise m the spectrum, then the needle 1snot spraying Repeat Subheading 3.3., step to open the needle tip If the spray becomes mstable, then Jitter and spikes will appear m the spectrum or the spectrum will contam an unusually htgh level of chemrcal background Increasing the air pressure of the needle helps stabilize the flow If thts sample measure 1sInsufficient, then the spraying voltage may be too low (increase tt by 100-150 V) or the openmg of the needle tip is still too small (repeat Subheading 3.3., step 8) Be aware that the nanoelectrospray needle has a very small diameter Small changes m the apphed potenttal (increases of 100 V) change the field density at the tip considerably Electrtcal dtscharge can be mmated and ionize atmospheric gas, whtch generates a mass spectrum These tons are usually small (~400 Dalton), and therefore, chemical background Ions m the higher mass region are mtssmg Atmospheric gas tomzatton can be vtsrble as a blue corona around the needle ttp (when light sources are switched off) and may lead to oxidations of methtonine-contaimng pepttdes (34) If opening of a needle IS not successful by the means described above, then apply voltage to the needle, pressurize it, and briefly touch rt against the mass spectrometer Interface plate (the potential drfference between needle and mterface plate should be about 500 V) The combined mechamcal and electrtcal stress opens almost every needle It is not advisable to open needles routmely by this approach, because tt tends to damage the metal coating at the tip High electrrcal current drawn from very thin tips can heat the glass to a degree that the glass melts The damaged piece of the tip can often be broken off, and analysts can proceed with a larger opening Two effects can prevent or stop spraying from an opened needle the high surface tensron of the liquid or preclpttatton of salts, polymers, or other nonvolattle sample constrtuents when then concentratton 1s high In the former case, the needle can easily be reopened by slightly touching tt to the mterface plate, thereby destroying the surface tenston by physrcal contact In the latter case, the harsher procedure employmg mechanical and electrical stress for needle openmg (described above) is perhaps required References Fenn, J B., Mann, M., Meng, C K., Wong, S F., and Whttehouse, C M (1989) Electrospray tomzatron for the mass spectrometry of large blomolecules Sczence 246,647 Covey, T R., Huang, E C , and Hemon, J D (1991) Structural charactertzatton of protein tryptic peptrdes via hqutd chromatography/mass spectrometry and collision-Induced drssociatton of then doubly charged molecular tons Anal Chem 63, 1193-1200 Peptide Sequencing 587 Griffin, P R., Coffman, J A., Hood, L E., and Yates, J R III (1991) Structural analysis of proteins by captllary HPLC electrospray tandem mass spectrometry Int J Mass Spectrom Ion Proc 111, 131-149 Hunt, D F., Henderson, R A., Shabanowitz, J., Sakaguchi, K., Michel, H., Sevilir, N., et al (1992) Characterization of peptides bound to the class I MHC molecule HLA-A2 by mass spectrometry Sczence 255, 1261-1263 Wilm, M S and Mann, M (1994) Electrospray and Taylor-Cone theory, Dole’s beam of macromolecules at last? Int J Mass Spectrom Ion Processes 136, 167-180 Wilm, M and Mann, M (1996) Analytical properties of the nano electrospray ion source Anal Chem 66, l-8 Wilm, M , Shevchenko, A., Houthaeve, T., Breit, S., Schweigerer, L , Fotsis, T , et al (1996) Femtomole sequencing of proteins from polyacrylamide gels by nano electrospray mass spectrometry Nature 379,466-469 Muzio, M., Chmnaiyan, A M., Ktschkel, F C., Rourke, K , Shevchenko, A., Ni, J., et al (1996) FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-mducmg signaling complex Cell 85, 17-827 Shevchenko, A., Jensen, N., PodteleJmkov, A V , Sagliocco, F , Wilm, M , Vorm, O., et al (1996) Lmkmg genome and proteome by mass spectrometry large scale identification of yeast proteins from two dimensional gels Proc Nat1 Acad Scl USA 93, 14,44O-14,445 10 Shevchenko, A., Wilm, M , and Mann, M (1997) Pepttde sequencing by mass spectrometry for homology searches and cloning of genes J Protein Chem 16(5), 48 l-490 11 Lingner, J., Hughes, T R., Shevchenko, A., Mann, M , Lundblad, V , and Cech, T R (1997) Reverse transcriptase motifs in the catalytic subunit of telomerase Sczence 276,56 l-567 12 Varga-Weuz, P D., Wilm, M., Bonte, E., Dumas, K., Mann, M., and Becker, P B (1997) Chromatm-remodellmg factor CHRAC contains the ATPases ISWI and topoisomerase II Nature 388, 598-601 13 Roepstorff, P and Fohlmann, J (1984) Proposal for a common nomenclature for sequence ions m mass spectra of peptides Blamed Mass Spectrom 11,60 14 Biemann, K (1988) Contributions of mass spectrometry to peptide and protein structure Blamed Enwronm Mass Spectrom 16,99-l 11 15 Hunt, D F., Yates, J R , Shabanowitz, J., Winston, S., and Hauer, C R (1986) Peptide Sequencing by Tandem Mass Spectrometry Proc Nat1 Acad Ser USA 83,6233-6237 16 Biemann, K and Scoble, H A (1987) Characterization by tandem mass spectrometry of structural modtficattons m proteins Science 237, 992 17 Wtlm, M , Neubauer, G., and Mann, M (1996) Parent ion scans of unseparated peptide mixtures Anal Chem 68,527-533 18 Huddleston, M J., Annan, R S., Bean, M T , and Carr, S A (1993) Selective detection of phosphopeptides in complex mixtures by electrospray liquid chromatography mass spectrometry J Am Sot Mass Spectrom 4,7 10-7 17 Jensen eta/ 588 19 Carr, S A., Huddleston, M J , and Annan, R S (1996) Selective detection and sequencing of phosphopeptrdes at the femtomole level by mass spectrometry Anal Biochem 239, 180-192 20 Carr, S A., Huddleston, M J., and Bean, M F (1993) Selective identification and differentiation of N- and U-linked oligosacchartdes m glycoproteins by liquid chromatography-mass spectrometry Protein Scz 2, 183-196 1, Neubauer, G and Mann, M (1997) Parent ion scans of large molecules J Mass Spectrom 32,94-98 22 Shevchenko, A., Chernuchevmh, I., Ens, W , Standmg, K , Thomson, B., Wilm, M., et al (1997) Rapid de novo pepttde sequencmg by a combmation of nanoelectrospray, Isotope labelmg and a quadrupole/time-of-flight mass spectrometer Rapzd Commun Mass Spectrom 11,lO 15-l 024 23 Mann, M and Wilm, M S (1994) Error tolerant identification of peptides in sequence databases by peptide sequence tags Anal Chem 66,439%4399 24 Mann, M (1994) Sequence database searchmg by mass spectrometric data, in Mzcrocharacterzzation ofProteins (Kellner, R , Lottspemh, F., and Meyer, H E., eds ), VCH Weinheim, pp 223-245 25 Jiang, X., Smith, J B., and Abraham, E C (1996) Identificatton of a MSMS diagnostic for methionine sulfoxide J Muss Spectrom 31, 1309-l 10 26 Lagerwerf, F M , van der Weert, M., Heerma, W , and Haverkamp, J (1996) Identification of oxidized methtomne m peptides Rapid Commun Mass Spectrom 10,1905-1910 27 Mann, M (1996) A shortcut to interesting human genes: Peptide Sequence Tags, ESTs and Computers Trends rn Blol Scz 21,494-495 28 Knapp, D R (1990) Chemical derivatization for mass spectrometry, m Meth Enzymol 193 (McCloskey, J A , ed ), (Academic, New York), pp 14-329 29 Takao, T., Hart, H , Okamoto, K , Harada, A., Kamachi, M , and Shimomshi, Y (1991) Facile assignment of sequence ions of a peptide labelled with I80 at the carboxyl terminus Raprd Commun Mass Spectrom 5,3 12-3 15 30 Schnolzer, M., Jedrzejewski, P , and Lehmann, W D (1996) Protease catalyzed mcorporation of 18-O mto peptide fragments and Its application for protein sequencing by electrospray and MALD iomzatron mass spectrometry Electrophoreszs 17 I Kruger, M S., Cook, K D., and Ramsey, R S (1995) Durable gold coated fused silica capillaries for use in electrospray mass spectrometty Anal Chem 67,385-389 32 Valaskovrc, G A and McLafferty, F W (1996) Long-lived metallized tips for nanohter electrospray mass spectrometry J Am Sot Mass Spectrom 7, 1270-l 272 33 Taylor, G (1964) Taylor cone theory Proc R Sot Lond Ser A 280,383 34 Morand, K., Talbo, G , and Mann, M (1993) Oxidation of peptides during electrospray iomzation Raprd Commun Mass Spectrom I, 738-743 ... Methods m Molecular Bdogy, Vol 112 2-D Proteome Analysrs Edlted by A J Lmk Humana Press Inc , Totowa, NJ Protocols Rabilloud Keeping proteins in solution during the 2-D electrophoresis process Although... Biology, Vol 112 2-D Proteome Analysrs Edlted by A J Lmk Humana Press tnc , Totowa, NJ 43 Protocols L willing 44 +- non linear IPG 3.510 4.44.64.85 5.2 5.4 5.6 5.8 6.2 0.4 6.6 Mr kDa Fig 1 .2-D gel electrophoresis... characterization by sequence analysis or mass spectrometry (Chapters 48-55) 2-D Protein Gel Electrophoresis t t Fig 2.A schematic showingthe SDS-PAGE usedfor the second gel dimension 2-D gel of electrophoresis